20181227.md

ArXiv cs.CV --Thu, 27 Dec 2018

1.A Poisson-Gaussian Denoising Dataset with Real Fluorescence Microscopy Images pdf

Fluorescence microscopy has enabled a dramatic development in modern biology. Due to its inherently weak signal, fluorescence microscopy is not only much noisier than photography, but also presented with Poisson-Gaussian noise where Poisson noise, or shot noise, is the dominating noise source, instead of Gaussian noise that dominates in photography. To get clean fluorescence microscopy images, it is highly desirable to have effective denoising algorithms and datasets that are specifically designed to denoise fluorescence microscopy images. While such algorithms exist, there are no such datasets available. In this paper, we fill this gap by constructing a dataset - the Fluorescence Microscopy Denoising (FMD) dataset - that is dedicated to Poisson-Gaussian denoising. The dataset consists 12,000 real fluorescence microscopy images obtained with commercial confocal, two-photon, and wide-field microscopes and representative biological samples such as cells, zebrafish, and mouse brain tissues. We use imaging averaging to effectively obtain ground truth images and 60,000 noisy images with different noise levels. We use this dataset to benchmark 10 representative denoising algorithms and find that deep learning methods have the best performance. To our knowledge, this is the first microscopy image dataset for Poisson-Gaussian denoising purposes and it could be an important tool for high-quality, real-time denoising applications in biomedical research.

2.Informative Object Annotations: Tell Me Something I Don't Know pdf

Capturing the interesting components of an image is a key aspect of image understanding. When a speaker annotates an image, selecting labels that are informative greatly depends on the prior knowledge of a prospective listener. Motivated by cognitive theories of categorization and communication, we present a new unsupervised approach to model this prior knowledge and quantify the informativeness of a description. Specifically, we compute how knowledge of a label reduces uncertainty over the space of labels and utilize this to rank candidate labels for describing an image. While the full estimation problem is intractable, we describe an efficient algorithm to approximate entropy reduction using a tree-structured graphical model. We evaluate our approach on the open-images dataset using a new evaluation set of 10K ground-truth ratings and find that it achieves ~65% agreement with human raters, largely outperforming other unsupervised baseline approaches.

3.Learning Not to Learn: Training Deep Neural Networks with Biased Data pdf

We propose a novel regularization algorithm to train deep neural networks, in which data at training time is severely biased. Since a neural network efficiently learns data distribution, a network is likely to learn the bias information to categorize input data. It leads to poor performance at test time, if the bias is, in fact, irrelevant to the categorization. In this paper, we formulate a regularization loss based on mutual information between feature embedding and bias. Based on the idea of minimizing this mutual information, we propose an iterative algorithm to unlearn the bias information. We employ an additional network to predict the bias distribution and train the network adversarially against the feature embedding network. At the end of learning, the bias prediction network is not able to predict the bias not because it is poorly trained, but because the feature embedding network successfully unlearns the bias information. We also demonstrate quantitative and qualitative experimental results which show that our algorithm effectively removes the bias information from feature embedding.

4.Region Proposal Networks with Contextual Selective Attention for Real-Time Organ Detection pdf

State-of-the-art methods for object detection use region proposal networks (RPN) to hypothesize object location. These networks simultaneously predicts object bounding boxes and \emph{objectness} scores at each location in the image. Unlike natural images for which RPN algorithms were originally designed, most medical images are acquired following standard protocols, thus organs in the image are typically at a similar location and possess similar geometrical characteristics (e.g. scale, aspect-ratio, etc.). Therefore, medical image acquisition protocols hold critical localization and geometric information that can be incorporated for faster and more accurate detection. This paper presents a novel attention mechanism for the detection of organs by incorporating imaging protocol information. Our novel selective attention approach (i) effectively shrinks the search space inside the feature map, (ii) appends useful localization information to the hypothesized proposal for the detection architecture to learn where to look for each organ, and (iii) modifies the pyramid of regression references in the RPN by incorporating organ- and modality-specific information, which results in additional time reduction. We evaluated the proposed framework on a dataset of 768 chest X-ray images obtained from a diverse set of sources. Our results demonstrate superior performance for the detection of the lung field compared to the state-of-the-art, both in terms of detection accuracy, demonstrating an improvement of $>7%$ in Dice score, and reduced processing time by $27.53%$ due to fewer hypotheses.

5.A Multi-Stream Convolutional Neural Network Framework for Group Activity Recognition pdf

In this work, we present a framework based on multi-stream convolutional neural networks (CNNs) for group activity recognition. Streams of CNNs are separately trained on different modalities and their predictions are fused at the end. Each stream has two branches to predict the group activity based on person and scene level representations. A new modality based on the human pose estimation is presented to add extra information to the model. We evaluate our method on the Volleyball and Collective Activity datasets. Experimental results show that the proposed framework is able to achieve state-of-the-art results when multiple or single frames are given as input to the model with 90.50% and 86.61% accuracy on Volleyball dataset, respectively, and 87.01% accuracy of multiple frames group activity on Collective Activity dataset.

6.Cluster Loss for Person Re-Identification pdf

Person re-identification (ReID) is an important problem in computer vision, especially for video surveillance applications. The problem focuses on identifying people across different cameras or across different frames of the same camera. The main challenge lies in identifying the similarity of the same person against large appearance and structure variations, while differentiating between individuals. Recently, deep learning networks with triplet loss have become a common framework for person ReID. However, triplet loss focuses on obtaining correct orders on the training set. We demonstrate that it performs inferior in a clustering task. In this paper, we design a cluster loss, which can lead to the model output with a larger inter-class variation and a smaller intra-class variation compared to the triplet loss. As a result, our model has a better generalization ability and can achieve higher accuracy on the test set especially for a clustering task. We also introduce a batch hard training mechanism for improving the results and faster convergence of training.

7.Structure-Aware 3D Hourglass Network for Hand Pose Estimation from Single Depth Image pdf

In this paper, we propose a novel structure-aware 3D hourglass network for hand pose estimation from a single depth image, which achieves state-of-the-art results on MSRA and NYU datasets. Compared to existing works that perform image-to-coordination regression, our network takes 3D voxel as input and directly regresses 3D heatmap for each joint. To be specific, we use hourglass network as our backbone network and modify it into 3D form. We explicitly model tree-like finger bone into the network as well as in the loss function in an end-to-end manner, in order to take the skeleton constraints into consideration. Final estimation can then be easily obtained from voxel density map with simple post-processing. Experimental results show that the proposed structure-aware 3D hourglass network is able to achieve a mean joint error of 7.4 mm in MSRA and 8.9 mm in NYU datasets, respectively.

8.Spotting Micro-Expressions on Long Videos Sequences pdf

This paper presents baseline results for the first Micro-Expression Spotting Challenge 2019 by evaluating local temporal pattern (LTP) on SAMM and CAS(ME)2. The proposed LTP patterns are extracted by applying PCA in a temporal window on several facial local regions. The micro-expression sequences are then spotted by a local classification of LTP and a global fusion. The performance is evaluated by Leave-One-Subject-Out cross validation. Furthermore, we define the criteria of determining true positives in one video by overlap rate and set the metric F1-score for spotting performance of the whole database. The F1-score of baseline results for SAMM and CAS(ME)2 are 0.0316 and 0.0179, respectively.

9.Spatial and Temporal Mutual Promotion for Video-based Person Re-identification pdf

Video-based person re-identification is a crucial task of matching video sequences of a person across multiple camera views. Generally, features directly extracted from a single frame suffer from occlusion, blur, illumination and posture changes. This leads to false activation or missing activation in some regions, which corrupts the appearance and motion representation. How to explore the abundant spatial-temporal information in video sequences is the key to solve this problem. To this end, we propose a Refining Recurrent Unit (RRU) that recovers the missing parts and suppresses noisy parts of the current frame's features by referring historical frames. With RRU, the quality of each frame's appearance representation is improved. Then we use the Spatial-Temporal clues Integration Module (STIM) to mine the spatial-temporal information from those upgraded features. Meanwhile, the multi-level training objective is used to enhance the capability of RRU and STIM. Through the cooperation of those modules, the spatial and temporal features mutually promote each other and the final spatial-temporal feature representation is more discriminative and robust. Extensive experiments are conducted on three challenging datasets, i.e., iLIDS-VID, PRID-2011 and MARS. The experimental results demonstrate that our approach outperforms existing state-of-the-art methods of video-based person re-identification on iLIDS-VID and MARS and achieves favorable results on PRID-2011.

10.A Survey to Deep Facial Attribute Analysis pdf

Facial attribute analysis has received considerable attention with the development of deep neural networks in the past few years. Facial attribute analysis contains two crucial issues: Facial Attribute Estimation (FAE), which recognizes whether facial attributes are present in given images, and Facial Attribute Manipulation (FAM), which synthesizes or removes desired facial attributes. In this paper, we provide a comprehensive survey on deep facial attribute analysis covering FAE and FAM. First, we present the basic knowledge of the two stages (i.e., data pre-processing and model construction) in the general deep facial attribute analysis pipeline. Second, we summarize the commonly used datasets and performance metrics. Third, we create a taxonomy of the state-of-the-arts and review detailed algorithms in FAE and FAM, respectively. Furthermore, we introduce several additional facial attribute related issues and applications. Finally, the possible challenges and future research directions are discussed.

11.A Whole Slide Image Grading Benchmark and Tissue Classification for Cervical Cancer Precursor Lesions with Inter-Observer Variability pdf

The cervical cancer developing from the precancerous lesions caused by the Human Papilloma Virus (HPV) has been one of the preventable cancers with the help of periodic screening. There are two types of grading conventions widely accepted among pathologists. On the other hand, inter-observer variability is an important issue for final diagnosis. In this paper, a whole-slide image grading benchmark for cervical cancer precursor lesions is introduced. The papillae of the cervical epithelium and overlapping cell problems are handled and a tissue classification method with a novel morphological feature exploiting the relative orientation between the BM and the major axis of all nuclei is developed and its performance is evaluated. Besides, the inter-observer variability is also revealed by a thorough comparison among pathologists' decisions, as well as, the final diagnosis.

12.3D PersonVLAD: Learning Deep Global Representations for Video-based Person Re-identification pdf

In this paper, we introduce a global video representation to video-based person re-identification (re-ID) that aggregates local 3D features across the entire video extent. Most of the existing methods rely on 2D convolutional networks (ConvNets) to extract frame-wise deep features which are pooled temporally to generate the video-level representations. However, 2D ConvNets lose temporal input information immediately after the convolution, and a separate temporal pooling is limited in capturing human motion in shorter sequences. To this end, we present a \textit{global} video representation (3D PersonVLAD), complementary to 3D ConvNets as a novel layer to capture the appearance and motion dynamics in full-length videos. However, encoding each video frame in its entirety and computing an aggregate global representation across all frames is tremendously challenging due to occlusions and misalignments. To resolve this, our proposed network is further augmented with 3D part alignment module to learn local features through soft-attention module. These attended features are statistically aggregated to yield identity-discriminative representations. Our global 3D features are demonstrated to achieve state-of-the-art results on three benchmark datasets: MARS \cite{MARS}, iLIDS-VID \cite{VideoRanking}, and PRID 2011

13.Practical Adversarial Attack Against Object Detector pdf

In this paper, we proposed the first practical adversarial attacks against object detectors in realistic situations: the adversarial examples are placed in different angles and distances, especially in the long distance (over 20m) and wide angles 120 degree. To improve the robustness of adversarial examples, we proposed the nested adversarial examples and introduced the image transformation techniques. Transformation methods aim to simulate the variance factors such as distances, angles, illuminations, etc., in the physical world. Two kinds of attacks were implemented on YOLO V3, a state-of-the-art real-time object detector: hiding attack that fools the detector unable to recognize the object, and appearing attack that fools the detector to recognize the non-existent object. The adversarial examples are evaluated in three environments: indoor lab, outdoor environment, and the real road, and demonstrated to achieve the success rate up to 92.4% based on the distance range from 1m to 25m. In particular, the real road testing of hiding attack on a straight road and a crossing road produced the success rate of 75% and 64% respectively, and the appearing attack obtained the success rates of 63% and 81% respectively, which we believe, should catch the attention of the autonomous driving community.

14.End-to-End Latent Fingerprint Search pdf

Latent fingerprints are one of the most important and widely used sources of evidence in law enforcement and forensic agencies. Yet the performance of the state-of-the-art latent recognition systems is far from satisfactory, and they often require manual markups to boost the latent search performance. Further, the COTS systems are proprietary and do not output the true comparison scores between a latent and reference prints to conduct quantitative evidential analysis. We present an end-to-end latent fingerprint search system, including automated region of interest (ROI) cropping, latent image preprocessing, feature extraction, feature comparison , and outputs a candidate list. Two separate minutiae extraction models provide complementary minutiae templates. To compensate for the small number of minutiae in small area and poor quality latents, a virtual minutiae set is generated to construct a texture template. A 96-dimensional descriptor is extracted for each minutia from its neighborhood. For computational efficiency, the descriptor length for virtual minutiae is further reduced to 16 using product quantization. Our end-to-end system is evaluated on three latent databases: NIST SD27 (258 latents); MSP (1,200 latents), WVU (449 latents) and N2N (10,000 latents) against a background set of 100K rolled prints, which includes the true rolled mates of the latents with rank-1 retrieval rates of 65.7%, 69.4%, 65.5%, and 7.6% respectively. A multi-core solution implemented on 24 cores obtains 1ms per latent to rolled comparison.

15.RegNet: Learning the Optimization of Direct Image-to-Image Pose Registration pdf

Direct image-to-image alignment that relies on the optimization of photometric error metrics suffers from limited convergence range and sensitivity to lighting conditions. Deep learning approaches has been applied to address this problem by learning better feature representations using convolutional neural networks, yet still require a good initialization. In this paper, we demonstrate that the inaccurate numerical Jacobian limits the convergence range which could be improved greatly using learned approaches. Based on this observation, we propose a novel end-to-end network, RegNet, to learn the optimization of image-to-image pose registration. By jointly learning feature representation for each pixel and partial derivatives that replace handcrafted ones (e.g., numerical differentiation) in the optimization step, the neural network facilitates end-to-end optimization. The energy landscape is constrained on both the feature representation and the learned Jacobian, hence providing more flexibility for the optimization as a consequence leads to more robust and faster convergence. In a series of experiments, including a broad ablation study, we demonstrate that RegNet is able to converge for large-baseline image pairs with fewer iterations.

16.FPD-M-net: Fingerprint Image Denoising and Inpainting Using M-Net Based Convolutional Neural Networks pdf

The fingerprint is a common biometric used for authentication and verification of an individual. These images are degraded when fingers are wet, dirty, dry or wounded and due to the failure of the sensors, etc. The extraction of the fingerprint from a degraded image requires denoising and inpainting. We propose to address these problems with an end-to-end trainable Convolutional Neural Network based architecture called FPD-M-net, by posing the fingerprint denoising and inpainting problem as a segmentation (foreground) task. Our architecture is based on the M-net with a change: structure similarity loss function, used for better extraction of the fingerprint from the noisy background. Our method outperforms the baseline method and achieves an overall 3rd rank in the Chalearn LAP Inpainting Competition Track 3 - Fingerprint Denoising and Inpainting, ECCV 2018

17.Deep Convolutional Generative Adversarial Network Based Food Recognition Using Partially Labeled Data pdf

Traditional machine learning algorithms using hand-crafted feature extraction techniques (such as local binary pattern) have limited accuracy because of high variation in images of the same class (or intra-class variation) for food recognition task. In recent works, convolutional neural networks (CNN) have been applied to this task with better results than all previously reported methods. However, they perform best when trained with large amount of annotated (labeled) food images. This is problematic when obtained in large volume, because they are expensive, laborious and impractical. Our work aims at developing an efficient deep CNN learning-based method for food recognition alleviating these limitations by using partially labeled training data on generative adversarial networks (GANs). We make new enhancements to the unsupervised training architecture introduced by Goodfellow et al. (2014), which was originally aimed at generating new data by sampling a dataset. In this work, we make modifications to deep convolutional GANs to make them robust and efficient for classifying food images. Experimental results on benchmarking datasets show the superiority of our proposed method as compared to the current-state-of-the-art methodologies even when trained with partially labeled training data.

18.Motion Selective Prediction for Video Frame Synthesis pdf

Existing conditional video prediction approaches train a network from large databases and generalize to previously unseen data. We take the opposite stance, and introduce a model that learns from the first frames of a given video and extends its content and motion, to, eg, double its length. To this end, we propose a dual network that can use in a flexible way both dynamic and static convolutional motion kernels, to predict future frames. The construct of our model gives us the the means to efficiently analyze its functioning and interpret its output. We demonstrate experimentally the robustness of our approach on challenging videos in-the-wild and show that it is competitive wrt related baselines.

19.The algorithm of the impulse noise filtration in images based on an algorithm of community detection in graphs pdf

This article suggests an algorithm of impulse noise filtration, based on the community detection in graphs. The image is representing as non-oriented weighted graph. Each pixel of an image is corresponding to a vertex of the graph. Community detection algorithm is running on the given graph. Assumed that communities that contain only one pixel are corresponding to noised pixels of an image. Suggested method was tested with help of computer experiment. This experiment was conducted on grayscale, and on colored images, on artificial images and on photos. It is shown that the suggested method is better than median filter by 20% regardless of noise percent. Higher efficiency is justified by the fact that most of filters are changing all of image pixels, but suggested method is finding and restoring only noised pixels. The dependence of the effectiveness of the proposed method on the percentage of noise in the image is shown.

20.Classification of X-Ray Protein Crystallization Using Deep Convolutional Neural Networks with a Finder Module pdf

Recently, deep convolutional neural networks have shown good results for image recognition. In this paper, we use convolutional neural networks with a finder module, which discovers the important region for recognition and extracts that region. We propose applying our method to the recognition of protein crystals for X-ray structural analysis. In this analysis, it is necessary to recognize states of protein crystallization from a large number of images. There are several methods that realize protein crystallization recognition by using convolutional neural networks. In each method, large-scale data sets are required to recognize with high accuracy. In our data set, the number of images is not good enough for training CNN. The amount of data for CNN is a serious issue in various fields. Our method realizes high accuracy recognition with few images by discovering the region where the crystallization drop exists. We compared our crystallization image recognition method with a high precision method using Inception-V3. We demonstrate that our method is effective for crystallization images using several experiments. Our method gained the AUC value that is about 5% higher than the compared method.

21.Adversarial Feature Genome: a Data Driven Adversarial Examples Recognition Method pdf

Convolutional neural networks (CNNs) are easily spoofed by adversarial examples which lead to wrong classification result. Most of the one-way defense methods focus only on how to improve the robustness of a CNN or to identify adversarial examples. They are incapable of identifying and correctly classifying adversarial examples simultaneously due to the lack of an effective way to quantitatively represent changes in the characteristics of the sample within the network. We find that adversarial examples and original ones have diverse representation in the feature space. Moreover, this difference grows as layers go deeper, which we call Adversarial Feature Separability (AFS). Inspired by AFS, we propose an Adversarial Feature Genome (AFG) based adversarial examples defense framework which can detect adversarial examples and correctly classify them into original category simultaneously. First, we extract the representations of adversarial examples and original ones with labels by the group visualization method. Then, we encode the representations into the feature database AFG. Finally, we model adversarial examples recognition as a multi-label classification or prediction problem by training a CNN for recognizing adversarial examples and original examples on the AFG. Experiments show that the proposed framework can not only effectively identify the adversarial examples in the defense process, but also correctly classify adversarial examples with mean accuracy up to 63%. Our framework potentially gives a new perspective, i.e. data-driven way, to adversarial examples defense. We believe that adversarial examples defense research may benefit from a large scale AFG database which is similar to ImageNet. The database and source code can be visited at this https URL.

22.Coupled Recurrent Network (CRN) pdf

Many semantic video analysis tasks can benefit from multiple, heterogenous signals. For example, in addition to the original RGB input sequences, sequences of optical flow are usually used to boost the performance of human action recognition in videos. To learn from these heterogenous input sources, existing methods reply on two-stream architectural designs that contain independent, parallel streams of Recurrent Neural Networks (RNNs). However, two-stream RNNs do not fully exploit the reciprocal information contained in the multiple signals, let alone exploit it in a recurrent manner. To this end, we propose in this paper a novel recurrent architecture, termed Coupled Recurrent Network (CRN), to deal with multiple input sources. In CRN, the parallel streams of RNNs are coupled together. Key design of CRN is a Recurrent Interpretation Block (RIB) that supports learning of reciprocal feature representations from multiple signals in a recurrent manner. Different from RNNs which stack the training loss at each time step or the last time step, we propose an effective and efficient training strategy for CRN. Experiments show the efficacy of the proposed CRN. In particular, we achieve the new state of the art on the benchmark datasets of human action recognition and multi-person pose estimation.

23.Selectivity or Invariance: Boundary-aware Salient Object Detection pdf

Typically, a salient object detection (SOD) model faces opposite requirements in processing object interiors and boundaries. The features of interiors should be invariant to strong appearance change so as to pop-out the salient object as a whole, while the features of boundaries should be selective to slight appearance change to distinguish salient objects and background. To address this selectivity-invariance dilemma, we propose a novel boundary-aware network with successive dilation for image-based SOD. In this network, the feature selectivity at boundaries is enhanced by incorporating a boundary localization stream, while the feature invariance at interiors is guaranteed with a complex interior perception stream. Moreover, a transition compensation stream is adopted to amend the probable failures in transitional regions between interiors and boundaries. In particular, an integrated successive dilation module is proposed to enhance the feature invariance at interiors and transitional regions. Extensive experiments on four datasets show that the proposed approach outperforms 11 state-of-the-art methods.

24.MMFNet: A Multi-modality MRI Fusion Network for Segmentation of Nasopharyngeal Carcinoma pdf

Segmentation of nasopharyngeal carcinoma (NPC) from Magnetic Resonance (MRI) Images is a crucial procedure for radiotherapy to improve clinical outcome and reduce radiation-associated toxicity. It is a time-consuming and label-intensive work for radiologists to manually mark the boundary of NPC slice by slice. In addition, due to the complex anatomical structure of NPC, automatic algorithms based on single-modality MRI do not have enough capability to get accurate delineation. To address the problem of weak distinction between normal adjacent tissues and lesion region in one single modality MRI, we propose a multi-modality MRI fusion network (MMFNet) to take advantage of three modalities MRI to realize NPC's segmentation. The backbone is a multi-encoder-based network, which is composed with several modality-specific encoders and one single decoder. The skip connection layer is utilized to combine low-level features from different modalities MRI with high-level features. Additionally, a fusion block is proposed to effectively fuse features from multi-modality MRI. Specifically speaking, the fusion block firstly highlight informative features and regions of interest, and then these weighted features will by fused and be further refined by a residual fusion block. Moreover, a training strategy named self-transfer is proposed to initializing encoders for multi-encoder-based network, it can stimulate encoders to make full mining of specific modality MRI. Our proposed framework can effectively make use of multi-modality medical datasets and the proposed modules such as fusion block and self-transfer can easily generalize to other multi-modality-based tasks.

25.Attention Branch Network: Learning of Attention Mechanism for Visual Explanation pdf

Visual explanation enables human to understand the decision making of Deep Convolutional Neural Network~(CNN), but it is insufficient to contribute the performance improvement. In this paper, we focus on the attention map for visual explanation, which represents high response value as the important region in image recognition. This region significantly improves the performance of CNN by introducing an attention mechanism that focuses on a specific region in an image. In this work, we propose Attention Branch Network~(ABN), which extends the top-down visual explanation model by introducing a branch structure with an attention mechanism. ABN can be applicable to several image recognition tasks by introducing a branch for attention mechanism and is trainable for the visual explanation and image recognition in end-to-end manner. We evaluate ABN on several image recognition tasks such as image classification, fine-grained recognition, and multiple facial attributes recognition. Experimental results show that ABN can outperform the accuracy of baseline models on these image recognition tasks while generating an attention map for visual explanation. Our code is availablethis https URL.

26.A Unified Framework for Mutual Improvement of SLAM and Semantic Segmentation pdf

This paper presents a novel framework for simultaneously implementing localization and segmentation, which are two of the most important vision-based tasks for robotics. While the goals and techniques used for them were considered to be different previously, we show that by making use of the intermediate results of the two modules, their performance can be enhanced at the same time. Our framework is able to handle both the instantaneous motion and long-term changes of instances in localization with the help of the segmentation result, which also benefits from the refined 3D pose information. We conduct experiments on various datasets, and prove that our framework works effectively on improving the precision and robustness of the two tasks and outperforms existing localization and segmentation algorithms.

27.Similarity R-C3D for Few-shot Temporal Activity Detection pdf

Many activities of interest are rare events, with only a few labeled examples available. Therefore models for temporal activity detection which are able to learn from a few examples are desirable. In this paper, we present a conceptually simple and general yet novel framework for few-shot temporal activity detection which detects the start and end time of the few-shot input activities in an untrimmed video. Our model is end-to-end trainable and can benefit from more few-shot examples. At test time, each proposal is assigned the label of the few-shot activity class corresponding to the maximum similarity score. Our Similarity R-C3D method outperforms previous work on three large-scale benchmarks for temporal activity detection (THUMOS14, ActivityNet1.2, and ActivityNet1.3 datasets) in the few-shot setting. Our code will be made available.

28.Curriculum Domain Adaptation for Semantic Segmentation of Urban Scenes pdf

During the last half decade, convolutional neural networks (CNNs) have triumphed over semantic segmentation, which is one of the core tasks in many applications such as autonomous driving and augmented reality. However, to train CNNs requires a considerable amount of data, which is difficult to collect and laborious to annotate. Recent advances in computer graphics make it possible to train CNNs on photo-realistic synthetic imagery with computer-generated annotations. Despite this, the domain mismatch between the real images and the synthetic data hinders the models' performance. Hence, we propose a curriculum-style learning approach to minimizing the domain gap in urban scene semantic segmentation. The curriculum domain adaptation solves easy tasks first to infer necessary properties about the target domain; in particular, the first task is to learn global label distributions over images and local distributions over landmark superpixels. These are easy to estimate because images of urban scenes have strong idiosyncrasies (e.g., the size and spatial relations of buildings, streets, cars, etc.). We then train a segmentation network, while regularizing its predictions in the target domain to follow those inferred properties. In experiments, our method outperforms the baselines on two datasets and two backbone networks. We also report extensive ablation studies about our approach.

29.Color Image Enhancement Method Based on Weighted Image Guided Filtering pdf

A novel color image enhancement method is proposed based on Retinex to enhance color images under non-uniform illumination or poor visibility conditions. Different from the conventional Retinex algorithms, the Weighted Guided Image Filter is used as a surround function instead of the Gaussian filter to estimate the background illumination, which can overcome the drawbacks of local blur and halo artifact that may appear by Gaussian filter. To avoid color distortion, the image is converted to the HSI color model, and only the intensity channel is enhanced. Then a linear color restoration algorithm is adopted to convert the enhanced intensity image back to the RGB color model, which ensures the hue is constant and undistorted. Experimental results show that the proposed method is effective to enhance both color and gray images with low exposure and non-uniform illumination, resulting in better visual quality than traditional method. At the same time, the objective evaluation indicators are also superior to the conventional methods. In addition, the efficiency of the proposed method is also improved thanks to the linear color restoration algorithm.

30.Image-to-Image Translation via Group-wise Deep Whitening and Coloring Transformation pdf

Unsupervised image translation is an active area powered by the advanced generative adversarial networks. Recently introduced models, such as DRIT or MUNIT, utilize a separate encoder in extracting the content and the style of image to successfully incorporate the multimodal nature of image translation. The existing methods, however, overlooks the role that the correlation between feature pairs plays in the overall style. The correlation between feature pairs on top of the mean and the variance of features, are important statistics that define the style of an image. In this regard, we propose an end-to-end framework tailored for image translation that leverages the covariance statistics by whitening the content of an input image followed by coloring to match the covariance statistics with an exemplar. The proposed group-wise deep whitening and coloring (GDWTC) algorithm is motivated by an earlier work of whitening and coloring transformation (WTC), but is augmented to be trained in an end-to-end manner, and with largely reduced computation costs. Our extensive qualitative and quantitative experiments demonstrate that the proposed GDWTC is fast, both in training and inference, and highly effective in reflecting the style of an exemplar.

31.Domain-Aware Generalized Zero-Shot Learning pdf

Generalized zero-shot learning (GZSL) is the problem of learning a classifier where some classes have samples, and others are learned from side information, like semantic attributes or text description, in a zero-shot learning fashion (ZSL). A major challenge in GZSL is to learn consistently for those two different domains. Here we describe a probabilistic approach that breaks the model into three modular components, and then combines them in a consistent way. Specifically, our model consists of three classifiers: A "gating" model that softly decides if a sample is from a "seen" class and two experts: a ZSL expert, and an expert model for seen classes. We address two main difficulties in this approach: How to provide an accurate estimate of the gating probability without any training samples for unseen classes; and how to use an expert predictions when it observes samples outside of its domain.
The key insight in our approach is to pass information between the three models to improve each others accuracy, while keeping the modular structure. We test our approach, Domain-Aware GZSL (DAZL) on three standard GZSL benchmark datasets (AWA, CUB, SUN), and find that it largely outperforms state-of-the-art GZSL models. DAZL is also the first model that closes the gap and surpasses the performance of generative models for GZSL, even-though it is a light-weight model that is much easier to train and tune.

32.TextNet: Irregular Text Reading from Images with an End-to-End Trainable Network pdf

Reading text from images remains challenging due to multi-orientation, perspective distortion and especially the curved nature of irregular text. Most of existing approaches attempt to solve the problem in two or multiple stages, which is considered to be the bottleneck to optimize the overall performance. To address this issue, we propose an end-to-end trainable network architecture, named TextNet, which is able to simultaneously localize and recognize irregular text from images. Specifically, we develop a scale-aware attention mechanism to learn multi-scale image features as a backbone network, sharing fully convolutional features and computation for localization and recognition. In text detection branch, we directly generate text proposals in quadrangles, covering oriented, perspective and curved text regions. To preserve text features for recognition, we introduce a perspective RoI transform layer, which can align quadrangle proposals into small feature maps. Furthermore, in order to extract effective features for recognition, we propose to encode the aligned RoI features by RNN into context information, combining spatial attention mechanism to generate text sequences. This overall pipeline is capable of handling both regular and irregular cases. Finally, text localization and recognition tasks can be jointly trained in an end-to-end fashion with designed multi-task loss. Experiments on standard benchmarks show that the proposed TextNet can achieve state-of-the-art performance, and outperform existing approaches on irregular datasets by a large margin.

33.Learning a Disentangled Embedding for Monocular 3D Shape Retrieval and Pose Estimation pdf

We propose a novel approach to jointly perform 3D object retrieval and pose estimation from monocular images.In order to make the method robust to real world scene variations in the images, e.g. texture, lighting and background,we learn an embedding space from 3D data that only includes the relevant information, namely the shape and pose.Our method can then be trained for robustness under real world scene variations without having to render a large training set simulating these variations. Our learned embedding explicitly disentangles a shape vector and a pose vector, which alleviates both pose bias for 3D shape retrieval and categorical bias for pose estimation. Having the learned disentangled embedding, we train a CNN to map the images to the embedding space, and then retrieve the closest 3D shape from the database and estimate the 6D pose of the object using the embedding vectors. Our method achieves 10.8 median error for pose estimation and 0.514 top-1-accuracy for category agnostic 3D object retrieval on the Pascal3D+ dataset. It therefore outperforms the previous state-of-the-art methods on both tasks.

34.Motion Blur removal via Coupled Autoencoder pdf

In this paper a joint optimization technique has been proposed for coupled autoencoder which learns the autoencoder weights and coupling map (between source and target) simultaneously. The technique is applicable to any transfer learning problem. In this work, we propose a new formulation that recasts deblurring as a transfer learning problem, it is solved using the proposed coupled autoencoder. The proposed technique can operate on-the-fly, since it does not require solving any costly inverse problem. Experiments have been carried out on state-of-the-art techniques, our method yields better quality images in shorter operating times.

35.Coupled Analysis Dictionary Learning to inductively learn inversion: Application to real-time reconstruction of Biomedical signals pdf

This work addresses the problem of reconstructing biomedical signals from their lower dimensional projections. Traditionally Compressed Sensing (CS) based techniques have been employed for this task. These are transductive inversion processes, the problem with these approaches is that the inversion is time-consuming and hence not suitable for real-time applications. With the recent advent of deep learning, Stacked Sparse Denoising Autoencoder (SSDAE) has been used for learning inversion in an inductive setup. The training period for inductive learning is large but is very fast during application -- capable of real-time speed. This work proposes a new approach for inductive learning of the inversion process. It is based on Coupled Analysis Dictionary Learning. Results on Biomedical signal reconstruction show that our proposed approach is very fast and yields result far better than CS and SSDAE.

36.Latent Filter Scaling for Multimodal Unsupervised Image-to-Image Translation pdf

In multimodal unsupervised image-to-image translation tasks, the goal is to translate an image from the source domain to many images in the target domain. We present a simple method that produces higher quality images than current state-of-the-art while maintaining the same amount of multimodal diversity. Previous methods follow the unconditional approach of trying to map the latent code directly to a full-size image. This leads to complicated network architectures with several introduced hyperparameters to tune. By treating the latent code as a modifier of the convolutional filters, we produce multimodal output while maintaining the traditional Generative Adversarial Network (GAN) loss and without additional hyperparameters. The only tuning required by our method controls the tradeoff between variability and quality of generated images. Furthermore, we achieve disentanglement between source domain content and target domain style for free as a by-product of our formulation. We perform qualitative and quantitative experiments showing the advantages of our method compared with the state-of-the art on multiple benchmark image-to-image translation datasets.

37.Perceptually-based single-image depth super-resolution pdf

RGBD images, combining high-resolution color and lower-resolution depth from various types of depth sensors, are increasingly common. One can significantly improve the resolution of depth images by taking advantage of color information; deep learning methods make combining color and depth information particularly easy. However, fusing these two sources of data may lead to a variety of artifacts. If depth maps are used to reconstruct 3D shapes, e.g., for virtual reality applications, the visual quality of upsampled images is particularly important. To achieve high-quality results, visual metric need to be taken into account. The main idea of our approach is to measure the quality of depth map upsampling using renderings of resulting 3D surfaces. We demonstrate that a simple visual appearance-based loss, when used with either a trained CNN or simply a deep prior, yields significantly improved 3D shapes, as measured by a number of existing perceptual metrics. We compare this approach with a number of existing optimization and learning-based techniques.

38.Holistic Decomposition Convolution for Effective Semantic Segmentation of 3D MR Images pdf

Convolutional Neural Networks (CNNs) have achieved state-of-the-art performance in many different 2D medical image analysis tasks. In clinical practice, however, a large part of the medical imaging data available is in 3D. This has motivated the development of 3D CNNs for volumetric image segmentation in order to benefit from more spatial context. Due to GPU memory restrictions caused by moving to fully 3D, state-of-the-art methods depend on subvolume/patch processing and the size of the input patch is usually small, limiting the incorporation of larger context information for a better performance. In this paper, we propose a novel Holistic Decomposition Convolution (HDC), for an effective and efficient semantic segmentation of volumetric images. HDC consists of a periodic down-shuffling operation followed by a conventional 3D convolution. HDC has the advantage of significantly reducing the size of the data for sub-sequential processing while using all the information available in the input irrespective of the down-shuffling factors. Results obtained from comprehensive experiments conducted on hip T1 MR images and intervertebral disc T2 MR images demonstrate the efficacy of the present approach.

39.Texture Deformation Based Generative Adversarial Networks for Face Editing pdf

Despite the significant success in image-to-image translation and latent representation based facial attribute editing and expression synthesis, the existing approaches still have limitations in the sharpness of details, distinct image translation and identity preservation. To address these issues, we propose a Texture Deformation Based GAN, namely TDB-GAN, to disentangle texture from original image and transfers domains based on the extracted texture. The approach utilizes the texture to transfer facial attributes and expressions without the consideration of the object pose. This leads to shaper details and more distinct visual effect of the synthesized faces. In addition, it brings the faster convergence during training. The effectiveness of the proposed method is validated through extensive ablation studies. We also evaluate our approach qualitatively and quantitatively on facial attribute and facial expression synthesis. The results on both the CelebA and RaFD datasets suggest that Texture Deformation Based GAN achieves better performance.

40.Precision Highway for Ultra Low-Precision Quantization pdf

Neural network quantization has an inherent problem called accumulated quantization error, which is the key obstacle towards ultra-low precision, e.g., 2- or 3-bit precision. To resolve this problem, we propose precision highway, which forms an end-to-end high-precision information flow while performing the ultra low-precision computation. First, we describe how the precision highway reduce the accumulated quantization error in both convolutional and recurrent neural networks. We also provide the quantitative analysis of the benefit of precision highway and evaluate the overhead on the state-of-the-art hardware accelerator. In the experiments, our proposed method outperforms the best existing quantization methods while offering 3-bit weight/activation quantization with no accuracy loss and 2-bit quantization with a 2.45 % top-1 accuracy loss in ResNet-50. We also report that the proposed method significantly outperforms the existing method in the 2-bit quantization of an LSTM for language modeling.

41.Writer-Aware CNN for Parsimonious HMM-Based Offline Handwritten Chinese Text Recognition pdf

Recently, the hybrid convolutional neural network hidden Markov model (CNN-HMM) has been introduced for offline handwritten Chinese text recognition (HCTR) and has achieved state-of-the-art performance. In a CNN-HMM system, a handwritten text line is modeled by a series of cascading HMMs, each representing one character, and the posterior distributions of HMM states are calculated by CNN. However, modeling each of the large vocabulary of Chinese characters with a uniform and fixed number of hidden states requires high memory and computational costs and makes the tens of thousands of HMM state classes confusing. Another key issue of CNN-HMM for HCTR is the diversified writing style, which leads to model strain and a significant performance decline for specific writers. To address these issues, we propose a writer-aware CNN based on parsimonious HMM (WCNN-PHMM). Validated on the ICDAR 2013 competition of CASIA-HWDB database, the more compact WCNN-PHMM of a 7360-class vocabulary can achieve a relative character error rate (CER) reduction of 16.6% over the conventional CNN-HMM without considering language modeling. Moreover, the state-tying results of PHMM explicitly show the information sharing among similar characters and the confusion reduction of tied state classes. Finally, we visualize the learned writer codes and demonstrate the strong relationship with the writing styles of different writers. To the best of our knowledge, WCNN-PHMM yields the best results on the ICDAR 2013 competition set, demonstrating its power when enlarging the size of the character vocabulary.

42.Deep Learning for Inferring the Surface Solar Irradiance from Sky Imagery pdf

We present a novel approach to perform ground-based estimation and prediction of the surface solar irradiance with the view to predicting photovoltaic energy production. We propose the use of mini-batch k-means clustering to extract features, referred to as per cluster number of pixels (PCNP), from sky images taken by a low-cost fish eye camera. These features are first used to classify the sky as clear or cloudy using a single hidden layer neural network; the classification accuracy achieves 99.7%. If the sky is classified as cloudy, we propose to use a deep neural network having as input features the PCNP to predict intra-hour variability of the solar irradiance. Toward this objective, in this paper, we focus on estimating the deep neural network model relating the PCNP features and the solar irradiance, which is an important step before performing the prediction task. The proposed deep learning-based estimation approach is shown to have an accuracy of 95%.

43.Leveraging Class Similarity to Improve Deep Neural Network Robustness pdf

Traditionally artificial neural networks (ANNs) are trained by minimizing the cross-entropy between a provided groundtruth delta distribution (encoded as one-hot vector) and the ANN's predictive softmax distribution. It seems, however, unacceptable to penalize networks equally for missclassification between classes. Confusing the class "Automobile" with the class "Truck" should be penalized less than confusing the class "Automobile" with the class "Donkey". To avoid such representation issues and learn cleaner classification boundaries in the network, this paper presents a variation of cross-entropy loss which depends not only on the sample class but also on a data-driven prior "class-similarity distribution" across the classes encoded in a matrix form. We explore learning the class-similarity distribution using a datadriven method and then show that by training with our modified similarity-driven loss, we obtain slightly better generalization performance over multiple architectures and datasets as well as improved performance on noisy testing scenarios.

44.End-to-end Learning for Graph Decomposition pdf

We propose a novel end-to-end trainable framework for the graph decomposition problem. The minimum cost multicut problem is first converted to an unconstrained binary cubic formulation where cycle consistency constraints are incorporated into the objective function. The new optimization problem can be viewed as a Conditional Random Field (CRF) in which the random variables are associated with the binary edge labels of the initial graph and the hard constraints are introduced in the CRF as high-order potentials. The parameters of a standard Neural Network and the fully differentiable CRF are optimized in an end-to-end manner. Furthermore, our method utilizes the cycle constraints as meta-supervisory signals during the learning of the deep feature representations by taking the dependencies between the output random variables into account. We present analyses of the end-to-end learned representations, showing the impact of the joint training, on the task of clustering images of MNIST. We also validate the effectiveness of our approach both for the feature learning and the final clustering on the challenging task of real-world multi-person pose estimation.

45.Advanced Image Processing for Astronomical Images pdf

Image Processing in Astronomy is a major field of research and involves a lot of techniques pertaining to improve analyzing the properties of the celestial objects or obtaining preliminary inference from the image data. In this paper, we provide a comprehensive case study of advanced image processing techniques applied to Astronomical Galaxy Images for improved analysis, accurate inferences and faster analysis.

46.Image Processing on IOPA Radiographs: A comprehensive case study on Apical Periodontitis pdf

With the recent advancements in Image Processing Techniques and development of new robust computer vision algorithms, new areas of research within Medical Diagnosis and Biomedical Engineering are picking up pace. This paper provides a comprehensive in-depth case study of Image Processing, Feature Extraction and Analysis of Apical Periodontitis diagnostic cases in IOPA (Intra Oral Peri-Apical) Radiographs, a common case in oral diagnostic pipeline. This paper provides a detailed analytical approach towards improving the diagnostic procedure with improved and faster results with higher accuracy targeting to eliminate True Negative and False Positive cases.

47.Chinese Herbal Recognition based on Competitive Attentional Fusion of Multi-hierarchies Pyramid Features pdf

Convolution neural netwotks (CNNs) are successfully applied in image recognition task. In this study, we explore the approach of automatic herbal recognition with CNNs and build the standard Chinese herbs datasets firstly. According to the characteristics of herbal images, we proposed the competitive attentional fusion pyramid networks to model the features of herbal image, which mdoels the relationship of feature maps from different levels, and re-weights multi-level channels with channel-wise attention mechanism. In this way, we can dynamically adjust the weight of feature maps from various layers, according to the visual characteristics of each herbal image. Moreover, we also introduce the spatial attention to recalibrate the misaligned features caused by sampling in features amalgamation. Extensive experiments are conducted on our proposed datasets and validate the superior performance of our proposed models. The Chinese herbs datasets will be released upon acceptance to facilitate the research of Chinese herbal recognition.

48.Estimation and Restoration of Compositional Degradation Using Convolutional Neural Networks pdf

Image restoration from a single image degradation type, such as blurring, hazing, random noise, and compression has been investigated for decades. However, image degradations in practice are often a mixture of several types of degradation. Such compositional degradations complicate restoration because they require the differentiation of different degradation types and levels. In this paper, we propose a convolutional neural network (CNN) model for estimating the degradation properties of a given degraded image. Furthermore, we introduce an image restoration CNN model that adopts the estimated degradation properties as its input. Experimental results show that the proposed degradation estimation model can successfully infer the degradation properties of compositionally degraded images. The proposed restoration model can restore degraded images by exploiting the estimated degradation properties and can achieve both blind and nonblind image restorations.

49.EgoReID: Person re-identification in Egocentric Videos Acquired by Mobile Devices with First-Person Point-of-View pdf

Widespread use of wearable cameras and recording devices such as cellphones have opened the door to a lot of interesting research in first-person Point-of-view (POV) videos (egocentric videos). In recent years, we have seen the performance of video-based person Re-Identification (ReID) methods improve considerably. However, with the influx of varying video domains, such as egocentric videos, it has become apparent that there are still many open challenges to be faced. These challenges are a result of factors such as poor video quality due to ego-motion, blurriness, severe changes in lighting conditions and perspective distortions. To facilitate the research towards conquering these challenges, this paper contributes a new, first-of-its-kind dataset called EgoReID. The dataset is captured using 3 mobile cellphones with non-overlapping field-of-view. It contains 900 IDs and around 10,200 tracks with a total of 176,000 detections. Moreover, for each video we also provide 12-sensor meta data. Directly applying current approaches to our dataset results in poor performance. Considering the unique nature of our dataset, we propose a new framework which takes advantage of both visual and sensor meta data to successfully perform Person ReID. In this paper, we propose to adopt human body region parsing to extract local features from different body regions and then employ 3D convolution to better encode temporal information of each sequence of body parts. In addition, we also employ sensor meta data to determine target's next camera and their estimated time of arrival, such that the search is only performed among tracks present in the predicted next camera around the estimated time. This considerably improves our ReID performance as it significantly reduces our search space.

50.The algorithm of formation of a training set for an artificial neural network for image segmentation pdf

This article suggests an algorithm of formation a training set for artificial neural network in case of image segmentation. The distinctive feature of this algorithm is that it using only one image for segmentation. The segmentation performs using three-layer perceptron. The main method of the segmentation is a method of region growing. Neural network is using for get a decision to include pixel into an area or not. Impulse noise is using for generation of a training set. Pixels damaged by noise are not related to the same region. Suggested method has been tested with help of computer experiment in automatic and interactive modes.

51.Temporal Hockey Action Recognition via Pose and Optical Flows pdf

Recognizing actions in ice hockey using computer vision poses challenges due to bulky equipment and inadequate image quality. A novel two-stream framework has been designed to improve action recognition accuracy for hockey using three main components. First, pose is estimated via the Part Affinity Fields model to extract meaningful cues from the player. Second, optical flow (using LiteFlowNet) is used to extract temporal features. Third, pose and optical flow streams are fused and passed to fully-connected layers to estimate the hockey player's action. A novel publicly available dataset named HARPET (Hockey Action Recognition Pose Estimation, Temporal) was created, composed of sequences of annotated actions and pose of hockey players including their hockey sticks as an extension of human body pose. Three contributions are recognized. (1) The novel two-stream architecture achieves 85% action recognition accuracy, with the inclusion of optical flows increasing accuracy by about 10%. (2) The unique localization of hand-held objects (e.g., hockey sticks) as part of pose increases accuracy by about 13%. (3) For pose estimation, a bigger and more general dataset, MSCOCO, is successfully used for transfer learning to a smaller and more specific dataset, HARPET, achieving a PCKh of 87%.

52.Dimensionality Reduction of Hyperspectral Imagery Based on Spatial-spectral Manifold Learning pdf

The graph embedding (GE) methods have been widely applied for dimensionality reduction of hyperspectral imagery (HSI). However, a major challenge of GE is how to choose proper neighbors for graph construction and explore the spatial information of HSI data. In this paper, we proposed an unsupervised dimensionality reduction algorithm termed spatial-spectral manifold reconstruction preserving embedding (SSMRPE) for HSI classification. At first, a weighted mean filter (WMF) is employed to preprocess the image, which aims to reduce the influence of background noise. According to the spatial consistency property of HSI, the SSMRPE method utilizes a new spatial-spectral combined distance (SSCD) to fuse the spatial structure and spectral information for selecting effective spatial-spectral neighbors of HSI pixels. Then, it explores the spatial relationship between each point and its neighbors to adjusts the reconstruction weights for improving the efficiency of manifold reconstruction. As a result, the proposed method can extract the discriminant features and subsequently improve the classification performance of HSI. The experimental results on PaviaU and Salinas hyperspectral datasets indicate that SSMRPE can achieve better classification accuracies in comparison with some state-of-the-art methods.

53.Disentangling Latent Space for VAE by Label Relevant/Irrelevant Dimensions pdf

VAE requires the standard Gaussian distribution as a prior in the latent space. Since all codes tend to follow the same prior, it often suffers the so-called "posterior collapse". To avoid this, this paper introduces the class specific distribution for the latent code. But different from CVAE, we present a method for disentangling the latent space into the label relevant and irrelevant dimensions, $\bm{\mathrm{z}}_s$ and $\bm{\mathrm{z}}_u$, for a single input. We apply two separated encoders to map the input into $\bm{\mathrm{z}}_s$ and $\bm{\mathrm{z}}_u$ respectively, and then give the concatenated code to the decoder to reconstruct the input. The label irrelevant code $\bm{\mathrm{z}}_u$ represent the common characteristics of all inputs, hence they are constrained by the standard Gaussian, and their encoder is trained in amortized variational inference way, like VAE. While $\bm{\mathrm{z}}_s$ is assumed to follow the Gaussian mixture distribution in which each component corresponds to a particular class. The parameters for the Gaussian components in $\bm{\mathrm{z}}_s$ encoder are optimized by the label supervision in a global stochastic way. In theory, we show that our method is actually equivalent to adding a KL divergence term on the joint distribution of $\bm{\mathrm{z}}_s$ and the class label $c$, and it can directly increase the mutual information between $\bm{\mathrm{z}}_s$ and the label $c$. Our model can also be extended to GAN by adding a discriminator in the pixel domain so that it produces high quality and diverse images.

54.Fully Automatic Segmentation of Sublingual Veins from Retrained U-Net Model for Few Near Infrared Images pdf

Sublingual vein is commonly used to diagnose the health status. The width of main sublingual veins gives information of the blood circulation. Therefore, it is necessary to segment the main sublingual veins from the tongue automatically. In general, the dataset in the medical field is small, which is a challenge for training the deep learning model. In order to train the model with a small data set, the proposed method for automatically segmenting the sublingual veins is to re-train U-net model with different sets of the limited number of labels for the same training images. With pre-knowledge of the segmentation, the loss of the trained model will be convergence easier. To improve the performance of the segmentation further, a novel strategy of data augmentation was utilized. The operation for masking output of the model with the input was randomly switched on or switched off in each training step. This approach will force the model to learn the contrast invariance and avoid overfitting. Images of dataset were taken with the developed device using eight near infrared LEDs. The final segmentation results were evaluated on the validation dataset by the IoU metric.

55.Dissociable neural representations of adversarially perturbed images in deep neural networks and the human brain pdf

Despite the remarkable similarities between deep neural networks (DNN) and the human brain as shown in previous studies, the fact that DNNs still fall behind humans in many visual tasks suggests that considerable differences still exist between the two systems. To probe their dissimilarities, we leverage adversarial noise (AN) and adversarial interference (AI) images that yield distinct recognition performance in a prototypical DNN (AlexNet) and human vision. The evoked activity by regular (RE) and adversarial images in both systems is thoroughly compared. We find that representational similarity between RE and adversarial images in the human brain resembles their perceptual similarity. However, such representation-perception association is disrupted in the DNN. Especially, the representational similarity between RE and AN images idiosyncratically increases from low- to high-level layers. Furthermore, forward encoding modeling reveals that the DNN-brain hierarchical correspondence proposed in previous studies only holds when the two systems process RE and AI images but not AN images. These results might be due to the deterministic modeling approach of current DNNs. Taken together, our results provide a complementary perspective on the comparison between DNNs and the human brain, and highlight the need to characterize their differences to further bridge artificial and human intelligence research.

56.Multi-Frame Super-Resolution Reconstruction with Applications to Medical Imaging pdf

The optical resolution of a digital camera is one of its most crucial parameters with broad relevance for consumer electronics, surveillance systems, remote sensing, or medical imaging. However, resolution is physically limited by the optics and sensor characteristics. In addition, practical and economic reasons often stipulate the use of out-dated or low-cost hardware. Super-resolution is a class of retrospective techniques that aims at high-resolution imagery by means of software. Multi-frame algorithms approach this task by fusing multiple low-resolution frames to reconstruct high-resolution images. This work covers novel super-resolution methods along with new applications in medical imaging.

57.Wireless Software Synchronization of Multiple Distributed Cameras pdf

We present a method for precisely time-synchronizing the capture of image sequences from a collection of smartphone cameras connected over WiFi. Our method is entirely software-based, has only modest hardware requirements, and achieves an accuracy of less than 250 microseconds on unmodified commodity hardware. It does not use image content and synchronizes cameras prior to capture. The algorithm operates in two stages. In the first stage, we designate one device as the leader and synchronize each client device's clock to it by estimating network delay. Once clocks are synchronized, the second stage initiates continuous image streaming, estimates the relative phase of image timestamps between each client and the leader, and shifts the streams into alignment. We quantitatively validate our results on a multi-camera rig imaging a high-precision LED array and qualitatively demonstrate significant improvements to multi-view stereo depth estimation and stitching of dynamic scenes. We plan to open-source an Android implementation of our system 'libsoftwaresync', potentially inspiring new types of collective capture applications.

58.An Empirical Analysis of Deep Audio-Visual Models for Speech Recognition pdf

In this project, we worked on speech recognition, specifically predicting individual words based on both the video frames and audio. Empowered by convolutional neural networks, the recent speech recognition and lip reading models are comparable to human level performance. We re-implemented and made derivations of the state-of-the-art model. Then, we conducted rich experiments including the effectiveness of attention mechanism, more accurate residual network as the backbone with pre-trained weights and the sensitivity of our model with respect to audio input with/without noise.

59.Cross Lingual Speech Emotion Recognition: Urdu vs. Western Languages pdf

Cross-lingual speech emotion recognition is an important task for practical applications. The performance of automatic speech emotion recognition systems degrades in cross-corpus scenarios, particularly in scenarios involving multiple languages or a previously unseen language such as Urdu for which limited or no data is available. In this study, we investigate the problem of cross-lingual emotion recognition for Urdu language and contribute URDU---the first ever spontaneous Urdu-language speech emotion database. Evaluations are performed using three different Western languages against Urdu and experimental results on different possible scenarios suggest various interesting aspects for designing more adaptive emotion recognition system for such limited languages. In results, selecting training instances of multiple languages can deliver comparable results to baseline and augmentation a fraction of testing language data while training can help to boost accuracy for speech emotion recognition. URDU data is publicly available for further research.

60.Urban-Rural Environmental Gradient in a Developing City: Testing ENVI GIS Functionality pdf

The research performs urban ecosystem analysis supported by ENVI GIS by integrated studies on land cover types and geospatial modeling of Taipei city. The paper deals with the role of anthropogenic pressure on the structure of the landscape and change of land cover types. Methods included assessment of the impact from anthropogenic activities on the natural ecosystems, evaluation of the rate and scale of landscape dynamics using remote sensing data and GIS. The research aims to assist environmentalists and city planners to evaluate strategies for specific objectives of urban development in Taiwan, China.

61.Studying the Plasticity in Deep Convolutional Neural Networks using Random Pruning pdf

Recently there has been a lot of work on pruning filters from deep convolutional neural networks (CNNs) with the intention of reducing computations.The key idea is to rank the filters based on a certain criterion (say, l1-norm) and retain only the top ranked filters. Once the low scoring filters are pruned away the remainder of the network is fine tuned and is shown to give performance comparable to the original unpruned network. In this work, we report experiments which suggest that the comparable performance of the pruned network is not due to the specific criterion chosen but due to the inherent plasticity of deep neural networks which allows them to recover from the loss of pruned filters once the rest of the filters are fine-tuned. Specifically we show counter-intuitive results wherein by randomly pruning 25-50% filters from deep CNNs we are able to obtain the same performance as obtained by using state-of-the-art pruning methods. We empirically validate our claims by doing an exhaustive evaluation with VGG-16 and ResNet-50. We also evaluate a real world scenario where a CNN trained on all 1000 ImageNet classes needs to be tested on only a small set of classes at test time (say, only animals). We create a new benchmark dataset from ImageNet to evaluate such class specific pruning and show that even here a random pruning strategy gives close to state-of-the-art performance. Unlike existing approaches which mainly focus on the task of image classification, in this work we also report results on object detection and image segmentation. We show that using a simple random pruning strategy we can achieve significant speed up in object detection (74% improvement in fps) while retaining the same accuracy as that of the original Faster RCNN model. Similarly we show that the performance of a pruned Segmentation Network (SegNet) is actually very similar to that of the original unpruned SegNet.

62.A Survey on Non-rigid 3D Shape Analysis pdf

Shape is an important physical property of natural and manmade 3D objects that characterizes their external appearances. Understanding differences between shapes and modeling the variability within and across shape classes, hereinafter referred to as \emph{shape analysis}, are fundamental problems to many applications, ranging from computer vision and computer graphics to biology and medicine. This chapter provides an overview of some of the recent techniques that studied the shape of 3D objects that undergo non-rigid deformations including bending and stretching. Recent surveys that covered some aspects such classification, retrieval, recognition, and rigid or nonrigid registration, focused on methods that use shape descriptors. Descriptors, however, provide abstract representations that do not enable the exploration of shape variability. In this chapter, we focus on recent techniques that treated the shape of 3D objects as points in some high dimensional space where paths describe deformations. Equipping the space with a suitable metric enables the quantification of the range of deformations of a given shape, which in turn enables (1) comparing and classifying 3D objects based on their shape, (2) computing smooth deformations, i.e. geodesics, between pairs of objects, and (3) modeling and exploring continuous shape variability in a collection of 3D models. This article surveys and classifies recent developments in this field, outlines fundamental issues, discusses their potential applications in computer vision and graphics, and highlights opportunities for future research. Our primary goal is to bridge the gap between various techniques that have been often independently proposed by different communities including mathematics and statistics, computer vision and graphics, and medical image analysis.

63.Learning based Facial Image Compression with Semantic Fidelity Metric pdf

Surveillance and security scenarios usually require high efficient facial image compression scheme for face recognition and identification. While either traditional general image codecs or special facial image compression schemes only heuristically refine codec separately according to face verification accuracy metric. We propose a Learning based Facial Image Compression (LFIC) framework with a novel Regionally Adaptive Pooling (RAP) module whose parameters can be automatically optimized according to gradient feedback from an integrated hybrid semantic fidelity metric, including a successfully exploration to apply Generative Adversarial Network (GAN) as metric directly in image compression scheme. The experimental results verify the framework's efficiency by demonstrating performance improvement of 71.41%, 48.28% and 52.67% bitrate saving separately over JPEG2000, WebP and neural network-based codecs under the same face verification accuracy distortion metric. We also evaluate LFIC's superior performance gain compared with latest specific facial image codecs. Visual experiments also show some interesting insight on how LFIC can automatically capture the information in critical areas based on semantic distortion metrics for optimized compression, which is quite different from the heuristic way of optimization in traditional image compression algorithms.

64.Dual Principal Component Pursuit: Probability Analysis and Efficient Algorithms pdf

Recent methods for learning a linear subspace from data corrupted by outliers are based on convex $\ell_1$ and nuclear norm optimization and require the dimension of the subspace and the number of outliers to be sufficiently small. In sharp contrast, the recently proposed Dual Principal Component Pursuit (DPCP) method can provably handle subspaces of high dimension by solving a non-convex $\ell_1$ optimization problem on the sphere. However, its geometric analysis is based on quantities that are difficult to interpret and are not amenable to statistical analysis. In this paper we provide a refined geometric analysis and a new statistical analysis that show that DPCP can tolerate as many outliers as the square of the number of inliers, thus improving upon other provably correct robust PCA methods. We also propose a scalable Projected Sub-Gradient Method method (DPCP-PSGM) for solving the DPCP problem and show it admits linear convergence even though the underlying optimization problem is non-convex and non-smooth. Experiments on road plane detection from 3D point cloud data demonstrate that DPCP-PSGM can be more efficient than the traditional RANSAC algorithm, which is one of the most popular methods for such computer vision applications.

65.Dynamic Runtime Feature Map Pruning pdf

High bandwidth requirements are an obstacle for accelerating the training and inference of deep neural networks. Most previous research focuses on reducing the size of kernel maps for inference. We analyze parameter sparsity of six popular convolutional neural networks - AlexNet, MobileNet, ResNet-50, SqueezeNet, TinyNet, and VGG16. Of the networks considered, those using ReLU (AlexNet, SqueezeNet, VGG16) contain a high percentage of 0-valued parameters and can be statically pruned. Networks with Non-ReLU activation functions in some cases may not contain any 0-valued parameters (ResNet-50, TinyNet). We also investigate runtime feature map usage and find that input feature maps comprise the majority of bandwidth requirements when depth-wise convolution and point-wise convolutions used. We introduce dynamic runtime pruning of feature maps and show that 10% of dynamic feature map execution can be removed without loss of accuracy. We then extend dynamic pruning to allow for values within an epsilon of zero and show a further 5% reduction of feature map loading with a 1% loss of accuracy in top-1.

66.Improving MMD-GAN Training with Repulsive Loss Function pdf

Generative adversarial nets (GANs) are widely used to learn the data sampling process and their performance may heavily depend on the loss functions, given a limited computational budget. This study revisits MMD-GAN that uses the maximum mean discrepancy (MMD) as the loss function for GAN and makes two contributions. First, we argue that the existing MMD loss function may discourage the learning of fine details in data as it attempts to contract the discriminator outputs of real data. To address this issue, we propose a repulsive loss function to actively learn the difference among the real data by simply rearranging the terms in MMD. Second, inspired by the hinge loss, we propose a bounded Gaussian kernel to stabilize the training of MMD-GAN with the repulsive loss function. The proposed methods are applied to the unsupervised image generation tasks on CIFAR-10, STL-10, CelebA, and LSUN bedroom datasets. Results show that the repulsive loss function significantly improves over the MMD loss at no additional computational cost and outperforms other representative loss functions. The proposed methods achieve an FID score of 16.21 on the CIFAR-10 dataset using a single DCGAN network and spectral normalization.

67.Guessing Smart: Biased Sampling for Efficient Black-Box Adversarial Attacks pdf

We consider adversarial examples in the black-box decision-based scenario. Here, an attacker has access to the final classification of a model, but not its parameters or softmax outputs. Most attacks for this scenario are based either on transferability, which is unreliable, or random sampling, which is extremely slow. Focusing on the latter, we propose to improve sampling-based attacks with prior beliefs about the target domain. We identify two such priors, image frequency and surrogate gradients, and discuss how to integrate them into a unified sampling procedure. We then formulate the Biased Boundary Attack, which achieves a drastic speedup over the original Boundary Attack. Finally, we demonstrate that our approach outperforms most state-of-the-art attacks in a query-limited scenario and is especially effective at breaking strong defenses: Our submission scored second place in the targeted attack track of the NeurIPS 2018 Adversarial Vision Challenge.

68.Multi-modal Learning with Prior Visual Relation Reasoning pdf

Visual relation reasoning is a central component in recent cross-modal analysis tasks, which aims at reasoning about the visual relationships between objects and their properties. These relationships convey rich semantics and help to enhance the visual representation for improving cross-modal analysis. Previous works have succeeded in designing strategies for modeling latent relations or rigid-categorized relations and achieving the lift of performance. However, this kind of methods leave out the ambiguity inherent in the relations because of the diverse relational semantics of different visual appearances. In this work, we explore to model relations by contextual-sensitive embeddings based on human priori knowledge. We novelly propose a plug-and-play relation reasoning module injected with the relation embeddings to enhance image encoder. Specifically, we design upgraded Graph Convolutional Networks (GCN) to utilize the information of relation embeddings and relation directionality between objects for generating relation-aware image representations. We demonstrate the effectiveness of the relation reasoning module by applying it to both Visual Question Answering (VQA) and Cross-Modal Information Retrieval (CMIR) tasks. Extensive experiments are conducted on VQA 2.0 and CMPlaces datasets and superior performance is reported when comparing with state-of-the-art work.

69.Image Embedding of PMU Data for Deep Learning towards Transient Disturbance Classification pdf

This paper presents a study on power grid disturbance classification by Deep Learning (DL). A real synchrophasor set composing of three different types of disturbance events from the Frequency Monitoring Network (FNET) is used. An image embedding technique called Gramian Angular Field is applied to transform each time series of event data to a two-dimensional image for learning. Two main DL algorithms, i.e. CNN (Convolutional Neural Network) and RNN (Recurrent Neural Network) are tested and compared with two widely used data mining tools, the Support Vector Machine and Decision Tree. The test results demonstrate the superiority of the both DL algorithms over other methods in the application of power system transient disturbance classification.

70.Multi-Step Prediction of Occupancy Grid Maps with Recurrent Neural Networks pdf

We investigate the multi-step prediction of the drivable space, represented by Occupancy Grid Maps (OGMs), for autonomous vehicles. Our motivation is that accurate multi-step prediction of the drivable space can efficiently improve path planning and navigation resulting in safe, comfortable and optimum paths in autonomous driving. We train a variety of Recurrent Neural Network (RNN) based architectures on the OGM sequences from the KITTI dataset. The results demonstrate significant improvement of the prediction accuracy using our proposed difference learning method, incorporating motion related features, over the state of the art. We remove the egomotion from the OGM sequences by transforming them into a common frame. Although in the transformed sequences the KITTI dataset is heavily biased toward static objects, by learning the difference between subsequent OGMs, our proposed method provides accurate prediction over both the static and moving objects.