1.Probabilistic Future Prediction for Video Scene Understanding ⬇️
We present a novel deep learning architecture for probabilistic future prediction from video. We predict the future semantics, geometry and motion of complex real-world urban scenes and use this representation to control an autonomous vehicle. This work is the first to jointly predict ego-motion, static scene, and the motion of dynamic agents in a probabilistic manner, which allows sampling consistent, highly probable futures from a compact latent space. Our model learns a representation from RGB video with a spatio-temporal convolutional module. The learned representation can be explicitly decoded to future semantic segmentation, depth, and optical flow, in addition to being an input to a learnt driving policy. To model the stochasticity of the future, we introduce a conditional variational approach which minimises the divergence between the present distribution (what could happen given what we have seen) and the future distribution (what we observe actually happens). During inference, diverse futures are generated by sampling from the present distribution.
2.Extending Maps with Semantic and Contextual Object Information for Robot Navigation: a Learning-Based Framework using Visual and Depth Cues ⬇️
This paper addresses the problem of building augmented metric representations of scenes with semantic information from RGB-D images. We propose a complete framework to create an enhanced map representation of the environment with object-level information to be used in several applications such as human-robot interaction, assistive robotics, visual navigation, or in manipulation tasks. Our formulation leverages a CNN-based object detector (Yolo) with a 3D model-based segmentation technique to perform instance semantic segmentation, and to localize, identify, and track different classes of objects in the scene. The tracking and positioning of semantic classes is done with a dictionary of Kalman filters in order to combine sensor measurements over time and then providing more accurate maps. The formulation is designed to identify and to disregard dynamic objects in order to obtain a medium-term invariant map representation. The proposed method was evaluated with collected and publicly available RGB-D data sequences acquired in different indoor scenes. Experimental results show the potential of the technique to produce augmented semantic maps containing several objects (notably doors). We also provide to the community a dataset composed of annotated object classes (doors, fire extinguishers, benches, water fountains) and their positioning, as well as the source code as ROS packages.
3.Harmonizing Transferability and Discriminability for Adapting Object Detectors ⬇️
Recent advances in adaptive object detection have achieved compelling results in virtue of adversarial feature adaptation to mitigate the distributional shifts along the detection pipeline. Whilst adversarial adaptation significantly enhances the transferability of feature representations, the feature discriminability of object detectors remains less investigated. Moreover, transferability and discriminability may come at a contradiction in adversarial adaptation given the complex combinations of objects and the differentiated scene layouts between domains. In this paper, we propose a Hierarchical Transferability Calibration Network (HTCN) that hierarchically (local-region/image/instance) calibrates the transferability of feature representations for harmonizing transferability and discriminability. The proposed model consists of three components: (1) Importance Weighted Adversarial Training with input Interpolation (IWAT-I), which strengthens the global discriminability by re-weighting the interpolated image-level features; (2) Context-aware Instance-Level Alignment (CILA) module, which enhances the local discriminability by capturing the underlying complementary effect between the instance-level feature and the global context information for the instance-level feature alignment; (3) local feature masks that calibrate the local transferability to provide semantic guidance for the following discriminative pattern alignment. Experimental results show that HTCN significantly outperforms the state-of-the-art methods on benchmark datasets.
4.Automatic Lesion Detection System (ALDS) for Skin Cancer Classification Using SVM and Neural Classifiers ⬇️
Technology aided platforms provide reliable tools in almost every field these days. These tools being supported by computational power are significant for applications that need sensitive and precise data analysis. One such important application in the medical field is Automatic Lesion Detection System (ALDS) for skin cancer classification. Computer aided diagnosis helps physicians and dermatologists to obtain a second opinion for proper analysis and treatment of skin cancer. Precise segmentation of the cancerous mole along with surrounding area is essential for proper analysis and diagnosis. This paper is focused towards the development of improved ALDS framework based on probabilistic approach that initially utilizes active contours and watershed merged mask for segmenting out the mole and later SVM and Neural Classifier are applied for the classification of the segmented mole. After lesion segmentation, the selected features are classified to ascertain that whether the case under consideration is melanoma or non-melanoma. The approach is tested for varying datasets and comparative analysis is performed that reflects the effectiveness of the proposed system.
5.Belief Propagation Reloaded: Learning BP-Layers for Labeling Problems ⬇️
It has been proposed by many researchers that combining deep neural networks with graphical models can create more efficient and better regularized composite models. The main difficulties in implementing this in practice are associated with a discrepancy in suitable learning objectives as well as with the necessity of approximations for the inference. In this work we take one of the simplest inference methods, a truncated max-product Belief Propagation, and add what is necessary to make it a proper component of a deep learning model: We connect it to learning formulations with losses on marginals and compute the backprop operation. This BP-Layer can be used as the final or an intermediate block in convolutional neural networks (CNNs), allowing us to design a hierarchical model composing BP inference and CNNs at different scale levels. The model is applicable to a range of dense prediction problems, is well-trainable and provides parameter-efficient and robust solutions in stereo, optical flow and semantic segmentation.
6.Semantic Pyramid for Image Generation ⬇️
We present a novel GAN-based model that utilizes the space of deep features learned by a pre-trained classification model. Inspired by classical image pyramid representations, we construct our model as a Semantic Generation Pyramid -- a hierarchical framework which leverages the continuum of semantic information encapsulated in such deep features; this ranges from low level information contained in fine features to high level, semantic information contained in deeper features. More specifically, given a set of features extracted from a reference image, our model generates diverse image samples, each with matching features at each semantic level of the classification model. We demonstrate that our model results in a versatile and flexible framework that can be used in various classic and novel image generation tasks. These include: generating images with a controllable extent of semantic similarity to a reference image, and different manipulation tasks such as semantically-controlled inpainting and compositing; all achieved with the same model, with no further training.
7.Pyramidal Edge-maps based Guided Thermal Super-resolution ⬇️
Thermal imaging is a robust sensing technique but its consumer applicability is limited by the high cost of thermal sensors. Nevertheless, low-resolution thermal cameras are relatively affordable and are also usually accompanied by a high-resolution visible-range camera. This visible-range image can be used as a guide to reconstruct a high-resolution thermal image using guided super-resolution(GSR) techniques. However, the difference in wavelength-range of the input images makes this task challenging. Improper processing can introduce artifacts such as blur and ghosting, mainly due to texture and content mismatch. To this end, we propose a novel algorithm for guided super-resolution that explicitly tackles the issue of texture-mismatch caused due to multimodality. We propose a two-stage network that combines information from a low-resolution thermal and a high-resolution visible image with the help of multi-level edge-extraction and integration. The first stage of our network extracts edge-maps from the visual image at different pyramidal levels and the second stage integrates these edge-maps into our proposed super-resolution network at appropriate layers. Extraction and integration of edges belonging to different scales simplifies the task of GSR as it provides texture to object-level information in a progressive manner. Using multi-level edges also allows us to adjust the contribution of the visual image directly at the time of testing and thus provides controllability at test-time. We perform multiple experiments and show that our method performs better than existing state-of-the-art guided super-resolution methods both quantitatively and qualitatively.
8.Adaptive Graph Convolutional Network with Attention Graph Clustering for Co-saliency Detection ⬇️
Co-saliency detection aims to discover the common and salient foregrounds from a group of relevant images. For this task, we present a novel adaptive graph convolutional network with attention graph clustering (GCAGC). Three major contributions have been made, and are experimentally shown to have substantial practical merits. First, we propose a graph convolutional network design to extract information cues to characterize the intra- and interimage correspondence. Second, we develop an attention graph clustering algorithm to discriminate the common objects from all the salient foreground objects in an unsupervised fashion. Third, we present a unified framework with encoder-decoder structure to jointly train and optimize the graph convolutional network, attention graph cluster, and co-saliency detection decoder in an end-to-end manner. We evaluate our proposed GCAGC method on three cosaliency detection benchmark datasets (iCoseg, Cosal2015 and COCO-SEG). Our GCAGC method obtains significant improvements over the state-of-the-arts on most of them.
9.Gimme Signals: Discriminative signal encoding for multimodal activity recognition ⬇️
We present a simple, yet effective and flexible method for action recognition supporting multiple sensor modalities. Multivariate signal sequences are encoded in an image and are then classified using a recently proposed EfficientNet CNN architecture. Our focus was to find an approach that generalizes well across different sensor modalities without specific adaptions while still achieving good results. We apply our method to 4 action recognition datasets containing skeleton sequences, inertial and motion capturing measurements as well as \wifi fingerprints that range up to 120 action classes. Our method defines the current best CNN-based approach on the NTU RGB+D 120 dataset, lifts the state of the art on the ARIL Wi-Fi dataset by +6.78%, improves the UTD-MHAD inertial baseline by +14.4%, the UTD-MHAD skeleton baseline by 1.13% and achieves 96.11% on the Simitate motion capturing data (80/20 split). We further demonstrate experiments on both, modality fusion on a signal level and signal reduction to prevent the representation from overloading.
10.PointINS: Point-based Instance Segmentation ⬇️
A single-point feature has shown its effectiveness in object detection. However, for instance segmentation, it does not lead to satisfactory results. The reasons are two folds. Firstly, it has limited representation capacity. Secondly, it could be misaligned with potential instances. To address the above issues, we propose a new point-based framework, namely PointINS, to segment instances from single points. The core module of our framework is instance-aware convolution, including the instance-agnostic feature and instance-aware weights. Instance-agnostic feature for each Point-of-Interest (PoI) serves as a template for potential instance masks. In this way, instance-aware features are computed by convolving this template with instance-aware weights for following mask prediction. Given the independence of instance-aware convolution, PointINS is general and practical as a one-stage detector for anchor-based and anchor-free frameworks. In our extensive experiments, we show the effectiveness of our framework on RetinaNet and FCOS. With ResNet101 backbone, PointINS achieves 38.3 mask mAP on challenging COCO dataset, outperforming its competitors by a large margin. The code will be made publicly available.
11.Is There Tradeoff between Spatial and Temporal in Video Super-Resolution? ⬇️
Recent advances of deep learning lead to great success of image and video super-resolution (SR) methods that are based on convolutional neural networks (CNN). For video SR, advanced algorithms have been proposed to exploit the temporal correlation between low-resolution (LR) video frames, and/or to super-resolve a frame with multiple LR frames. These methods pursue higher quality of super-resolved frames, where the quality is usually measured frame by frame in e.g. PSNR. However, frame-wise quality may not reveal the consistency between frames. If an algorithm is applied to each frame independently (which is the case of most previous methods), the algorithm may cause temporal inconsistency, which can be observed as flickering. It is a natural requirement to improve both frame-wise fidelity and between-frame consistency, which are termed spatial quality and temporal quality, respectively. Then we may ask, is a method optimized for spatial quality also optimized for temporal quality? Can we optimize the two quality metrics jointly?
12.Partial Weight Adaptation for Robust DNN Inference ⬇️
Mainstream video analytics uses a pre-trained DNN model with an assumption that inference input and training data follow the same probability distribution. However, this assumption does not always hold in the wild: autonomous vehicles may capture video with varying brightness; unstable wireless bandwidth calls for adaptive bitrate streaming of video; and, inference servers may serve inputs from heterogeneous IoT devices/cameras. In such situations, the level of input distortion changes rapidly, thus reshaping the probability distribution of the input.
We present GearNN, an adaptive inference architecture that accommodates heterogeneous DNN inputs. GearNN employs an optimization algorithm to identify a small set of "distortion-sensitive" DNN parameters, given a memory budget. Based on the distortion level of the input, GearNN then adapts only the distortion-sensitive parameters, while reusing the rest of constant parameters across all input qualities. In our evaluation of DNN inference with dynamic input distortions, GearNN improves the accuracy (mIoU) by an average of 18.12% over a DNN trained with the undistorted dataset and 4.84% over stability training from Google, with only 1.8% extra memory overhead.
13.Dual Temporal Memory Network for Efficient Video Object Segmentation ⬇️
Video Object Segmentation (VOS) is typically formulated in a semi-supervised setting. Given the ground-truth segmentation mask on the first frame, the task of VOS is to track and segment the single or multiple objects of interests in the rest frames of the video at the pixel level. One of the fundamental challenges in VOS is how to make the most use of the temporal information to boost the performance. We present an end-to-end network which stores short- and long-term video sequence information preceding the current frame as the temporal memories to address the temporal modeling in VOS. Our network consists of two temporal sub-networks including a short-term memory sub-network and a long-term memory sub-network. The short-term memory sub-network models the fine-grained spatial-temporal interactions between local regions across neighboring frames in video via a graph-based learning framework, which can well preserve the visual consistency of local regions over time. The long-term memory sub-network models the long-range evolution of object via a Simplified-Gated Recurrent Unit (S-GRU), making the segmentation be robust against occlusions and drift errors. In our experiments, we show that our proposed method achieves a favorable and competitive performance on three frequently-used VOS datasets, including DAVIS 2016, DAVIS 2017 and Youtube-VOS in terms of both speed and accuracy.
14.BIHL:A Fast and High Performance Object Proposals based on Binarized HL Frequency ⬇️
In recent years, the use of object proposal as a preprocessing step for target detection to improve computational efficiency has become an effective method. Good object proposal methods should have high object detection recall rate and low computational cost, as well as good location accuracy and repeatability. However, it is difficult for current advanced algorithms to achieve a good balance in the above performance. Therefore, it is especially important to ensure that the recall rate and location quality are not degraded while accelerating object generation.For this problem, we propose a class-independent object proposal algorithm BIHL. It combines the advantages of window scoring and superpixel merging. First, a binarized horizontal high frequency component feature and a linear classifier are used to learn and generate a set of candidate boxs with a objective score. Then, the candidate boxs are merged based on the principle of location and score proximity. Different from superpixel merging algorithm, our method does not use pixel level operation to avoid a lot of computation without losing performance. Experimental results on the VOC2007 dataset and the VOC2007 synthetic interference dataset containing 297,120 test images show that when including difficult-to-identify objects with an IOU threshold of 0.5 and 10000 budget proposals, our method achieves a 99.3% detection recall and an mean average best overlap of 81.1% . The average processing time of our method for all test set images is 0.0015 seconds, which is nearly 3 times faster than the current fastest method. In repeatability testing, our method is the method with the highest average repeatability among the methods that achieve good repeatability to various disturbances, and the average repeatability is 10% higher than RPN. The code will be published in {this https URL}
15.A Spatial-Temporal Attentive Network with Spatial Continuity for Trajectory Prediction ⬇️
It remains challenging to automatically predict the multi-agent trajectory due to multiple interactions including agent to agent interaction and scene to agent interaction. Although recent methods have achieved promising performance, most of them just consider spatial influence of the interactions and ignore the fact that temporal influence always accompanies spatial influence. Moreover, those methods based on scene information always require extra segmented scene images to generate multiple socially acceptable trajectories. To solve these limitations, we propose a novel model named spatial-temporal attentive network with spatial continuity (STAN-SC). First, spatial-temporal attention mechanism is presented to explore the most useful and important information. Second, we conduct a joint feature sequence based on the sequence and instant state information to make the generative trajectories keep spatial continuity. Experiments are performed on the two widely used ETH-UCY datasets and demonstrate that the proposed model achieves state-of-the-art prediction accuracy and handles more complex scenarios.
16.BigGAN-based Bayesian reconstruction of natural images from human brain activity ⬇️
In the visual decoding domain, visually reconstructing presented images given the corresponding human brain activity monitored by functional magnetic resonance imaging (fMRI) is difficult, especially when reconstructing viewed natural images. Visual reconstruction is a conditional image generation on fMRI data and thus generative adversarial network (GAN) for natural image generation is recently introduced for this task. Although GAN-based methods have greatly improved, the fidelity and naturalness of reconstruction are still unsatisfactory due to the small number of fMRI data samples and the instability of GAN training. In this study, we proposed a new GAN-based Bayesian visual reconstruction method (GAN-BVRM) that includes a classifier to decode categories from fMRI data, a pre-trained conditional generator to generate natural images of specified categories, and a set of encoding models and evaluator to evaluate generated images. GAN-BVRM employs the pre-trained generator of the prevailing BigGAN to generate masses of natural images, and selects the images that best matches with the corresponding brain activity through the encoding models as the reconstruction of the image stimuli. In this process, the semantic and detailed contents of reconstruction are controlled by decoded categories and encoding models, respectively. GAN-BVRM used the Bayesian manner to avoid contradiction between naturalness and fidelity from current GAN-based methods and thus can improve the advantages of GAN. Experimental results revealed that GAN-BVRM improves the fidelity and naturalness, that is, the reconstruction is natural and similar to the presented image stimuli.
17.Deep Domain-Adversarial Image Generation for Domain Generalisation ⬇️
Machine learning models typically suffer from the domain shift problem when trained on a source dataset and evaluated on a target dataset of different distribution. To overcome this problem, domain generalisation (DG) methods aim to leverage data from multiple source domains so that a trained model can generalise to unseen domains. In this paper, we propose a novel DG approach based on \emph{Deep Domain-Adversarial Image Generation} (DDAIG). Specifically, DDAIG consists of three components, namely a label classifier, a domain classifier and a domain transformation network (DoTNet). The goal for DoTNet is to map the source training data to unseen domains. This is achieved by having a learning objective formulated to ensure that the generated data can be correctly classified by the label classifier while fooling the domain classifier. By augmenting the source training data with the generated unseen domain data, we can make the label classifier more robust to unknown domain changes. Extensive experiments on four DG datasets demonstrate the effectiveness of our approach.
18.Interaction Graphs for Object Importance Estimation in On-road Driving Videos ⬇️
A vehicle driving along the road is surrounded by many objects, but only a small subset of them influence the driver's decisions and actions. Learning to estimate the importance of each object on the driver's real-time decision-making may help better understand human driving behavior and lead to more reliable autonomous driving systems. Solving this problem requires models that understand the interactions between the ego-vehicle and the surrounding objects. However, interactions among other objects in the scene can potentially also be very helpful, e.g., a pedestrian beginning to cross the road between the ego-vehicle and the car in front will make the car in front less important. We propose a novel framework for object importance estimation using an interaction graph, in which the features of each object node are updated by interacting with others through graph convolution. Experiments show that our model outperforms state-of-the-art baselines with much less input and pre-processing.
19.Analyzing Visual Representations in Embodied Navigation Tasks ⬇️
Recent advances in deep reinforcement learning require a large amount of training data and generally result in representations that are often over specialized to the target task. In this work, we present a methodology to study the underlying potential causes for this specialization. We use the recently proposed projection weighted Canonical Correlation Analysis (PWCCA) to measure the similarity of visual representations learned in the same environment by performing different tasks.
We then leverage our proposed methodology to examine the task dependence of visual representations learned on related but distinct embodied navigation tasks. Surprisingly, we find that slight differences in task have no measurable effect on the visual representation for both SqueezeNet and ResNet architectures. We then empirically demonstrate that visual representations learned on one task can be effectively transferred to a different task.
20.LaserFlow: Efficient and Probabilistic Object Detection and Motion Forecasting ⬇️
In this work, we present LaserFlow, an efficient method for 3D object detection and motion forecasting from LiDAR. Unlike the previous work, our approach utilizes the native range view representation of the LiDAR, which enables our method to operate at the full range of the sensor in real-time without voxelization or compression of the data. We propose a new multi-sweep fusion architecture, which extracts and merges temporal features directly from the range images. Furthermore, we propose a novel technique for learning a probability distribution over future trajectories inspired by curriculum learning. We evaluate LaserFlow on two autonomous driving datasets and demonstrate competitive results when compared to the existing state-of-the-art methods.
21.Sparse Graphical Memory for Robust Planning ⬇️
To operate effectively in the real world, artificial agents must act from raw sensory input such as images and achieve diverse goals across long time-horizons. On the one hand, recent strides in deep reinforcement and imitation learning have demonstrated impressive ability to learn goal-conditioned policies from high-dimensional image input, though only for short-horizon tasks. On the other hand, classical graphical methods like A* search are able to solve long-horizon tasks, but assume that the graph structure is abstracted away from raw sensory input and can only be constructed with task-specific priors. We wish to combine the strengths of deep learning and classical planning to solve long-horizon tasks from raw sensory input. To this end, we introduce Sparse Graphical Memory (SGM), a new data structure that stores observations and feasible transitions in a sparse memory. SGM can be combined with goal-conditioned RL or imitative agents to solve long-horizon tasks across a diverse set of domains. We show that SGM significantly outperforms current state of the art methods on long-horizon, sparse-reward visual navigation tasks. Project video and code are available at this https URL
22.Advanced Deep Learning Methodologies for Skin Cancer Classification in Prodromal Stages ⬇️
Technology-assisted platforms provide reliable solutions in almost every field these days. One such important application in the medical field is the skin cancer classification in preliminary stages that need sensitive and precise data analysis. For the proposed study the Kaggle skin cancer dataset is utilized. The proposed study consists of two main phases. In the first phase, the images are preprocessed to remove the clutters thus producing a refined version of training images. To achieve that, a sharpening filter is applied followed by a hair removal algorithm. Different image quality measurement metrics including Peak Signal to Noise (PSNR), Mean Square Error (MSE), Maximum Absolute Squared Deviation (MXERR) and Energy Ratio/ Ratio of Squared Norms (L2RAT) are used to compare the overall image quality before and after applying preprocessing operations. The results from the aforementioned image quality metrics prove that image quality is not compromised however it is upgraded by applying the preprocessing operations. The second phase of the proposed research work incorporates deep learning methodologies that play an imperative role in accurate, precise and robust classification of the lesion mole. This has been reflected by using two state of the art deep learning models: Inception-v3 and MobileNet. The experimental results demonstrate notable improvement in train and validation accuracy by using the refined version of images of both the networks, however, the Inception-v3 network was able to achieve better validation accuracy thus it was finally selected to evaluate it on test data. The final test accuracy using state of art Inception-v3 network was 86%.
23.Human Activity Recognition using Multi-Head CNN followed by LSTM ⬇️
This study presents a novel method to recognize human physical activities using CNN followed by LSTM. Achieving high accuracy by traditional machine learning algorithms, (such as SVM, KNN and random forest method) is a challenging task because the data acquired from the wearable sensors like accelerometer and gyroscope is a time-series data. So, to achieve high accuracy, we propose a multi-head CNN model comprising of three CNNs to extract features for the data acquired from different sensors and all three CNNs are then merged, which are followed by an LSTM layer and a dense layer. The configuration of all three CNNs is kept the same so that the same number of features are obtained for every input to CNN. By using the proposed method, we achieve state-of-the-art accuracy, which is comparable to traditional machine learning algorithms and other deep neural network algorithms.
24.A Power-Efficient Binary-Weight Spiking Neural Network Architecture for Real-Time Object Classification ⬇️
Neural network hardware is considered an essential part of future edge devices. In this paper, we propose a binary-weight spiking neural network (BW-SNN) hardware architecture for low-power real-time object classification on edge platforms. This design stores a full neural network on-chip, and hence requires no off-chip bandwidth. The proposed systolic array maximizes data reuse for a typical convolutional layer. A 5-layer convolutional BW-SNN hardware is implemented in 90nm CMOS. Compared with state-of-the-art designs, the area cost and energy per classification are reduced by 7$\times$ and 23$\times$, respectively, while also achieving a higher accuracy on the MNIST benchmark. This is also a pioneering SNN hardware architecture that supports advanced CNN architectures.
25.What Information Does a ResNet Compress? ⬇️
The information bottleneck principle (Shwartz-Ziv & Tishby, 2017) suggests that SGD-based training of deep neural networks results in optimally compressed hidden layers, from an information theoretic perspective. However, this claim was established on toy data. The goal of the work we present here is to test whether the information bottleneck principle is applicable to a realistic setting using a larger and deeper convolutional architecture, a ResNet model. We trained PixelCNN++ models as inverse representation decoders to measure the mutual information between hidden layers of a ResNet and input image data, when trained for (1) classification and (2) autoencoding. We find that two stages of learning happen for both training regimes, and that compression does occur, even for an autoencoder. Sampling images by conditioning on hidden layers' activations offers an intuitive visualisation to understand what a ResNets learns to forget.
26.Unsupervised Style and Content Separation by Minimizing Mutual Information for Speech Synthesis ⬇️
We present a method to generate speech from input text and a style vector that is extracted from a reference speech signal in an unsupervised manner, i.e., no style annotation, such as speaker information, is required. Existing unsupervised methods, during training, generate speech by computing style from the corresponding ground truth sample and use a decoder to combine the style vector with the input text. Training the model in such a way leaks content information into the style vector. The decoder can use the leaked content and ignore some of the input text to minimize the reconstruction loss. At inference time, when the reference speech does not match the content input, the output may not contain all of the content of the input text. We refer to this problem as "content leakage", which we address by explicitly estimating and minimizing the mutual information between the style and the content through an adversarial training formulation. We call our method MIST - Mutual Information based Style Content Separation. The main goal of the method is to preserve the input content in the synthesized speech signal, which we measure by the word error rate (WER) and show substantial improvements over state-of-the-art unsupervised speech synthesis methods.
27.On the effectiveness of convolutional autoencoders on image-based personalized recommender systems ⬇️
Recommender systems (RS) are increasingly present in our daily lives, especially since the advent of Big Data, which allows for storing all kinds of information about users' preferences. Personalized RS are successfully applied in platforms such as Netflix, Amazon or YouTube. However, they are missing in gastronomic platforms such as TripAdvisor, where moreover we can find millions of images tagged with users' tastes. This paper explores the potential of using those images as sources of information for modeling users' tastes and proposes an image-based classification system to obtain personalized recommendations, using a convolutional autoencoder as feature extractor. The proposed architecture will be applied to TripAdvisor data, using users' reviews that can be defined as a triad composed by a user, a restaurant, and an image of it taken by the user. Since the dataset is highly unbalanced, the use of data augmentation on the minority class is also considered in the experimentation. Results on data from three cities of different sizes (Santiago de Compostela, Barcelona and New York) demonstrate the effectiveness of using a convolutional autoencoder as feature extractor, instead of the standard deep features computed with convolutional neural networks.
28.Towards a Framework for Visual Intelligence in Service Robotics: Epistemic Requirements and Gap Analysis ⬇️
A key capability required by service robots operating in real-world, dynamic environments is that of Visual Intelligence, i.e., the ability to use their vision system, reasoning components and background knowledge to make sense of their environment. In this paper, we analyze the epistemic requirements for Visual Intelligence, both in a top-down fashion, using existing frameworks for human-like Visual Intelligence in the literature, and from the bottom up, based on the errors emerging from object recognition trials in a real-world robotic scenario. Finally, we use these requirements to evaluate current knowledge bases for Service Robotics and to identify gaps in the support they provide for Visual Intelligence. These gaps provide the basis of a research agenda for developing more effective knowledge representations for Visual Intelligence.
29.Random smooth gray value transformations for cross modality learning with gray value invariant networks ⬇️
Random transformations are commonly used for augmentation of the training data with the goal of reducing the uniformity of the training samples. These transformations normally aim at variations that can be expected in images from the same modality. Here, we propose a simple method for transforming the gray values of an image with the goal of reducing cross modality differences. This approach enables segmentation of the lumbar vertebral bodies in CT images using a network trained exclusively with MR images. The source code is made available at this https URL
30.LIBRE: The Multiple 3D LiDAR Dataset ⬇️
In this work, we present LIBRE: LiDAR Benchmarking and Reference, a first-of-its-kind dataset featuring 12 different LiDAR sensors, covering a range of manufacturers, models, and laser configurations. Data captured independently from each sensor includes four different environments and configurations: static obstacles placed at known distances and measured from a fixed position within a controlled environment; static obstacles measured from a moving vehicle, captured in a weather chamber where LiDARs were exposed to different conditions (fog, rain, strong light); dynamic objects actively measured from a fixed position by multiple LiDARs mounted side-by-side simultaneously, creating indirect interference conditions; and dynamic traffic objects captured from a vehicle driven on public urban roads multiple times at different times of the day, including data from supporting sensors such as cameras, infrared imaging, and odometry devices. LIBRE will contribute the research community to (1) provide a means for a fair comparison of currently available LiDARs, and (2) facilitate the improvement of existing self-driving vehicles and robotics-related software, in terms of development and tuning of LiDAR-based perception algorithms.
31.Action for Better Prediction ⬇️
Good prediction is necessary for autonomous robotics to make informed decisions in dynamic environments. Improvements can be made to the performance of a given data-driven prediction model by using better sampling strategies when collecting training data. Active learning approaches to optimal sampling have been combined with the mathematically general approaches to incentivizing exploration presented in the curiosity literature via model-based formulations of curiosity. We present an adversarial curiosity method which maximizes a score given by a discriminator network. This score gives a measure of prediction certainty enabling our approach to sample sequences of observations and actions which result in outcomes considered the least realistic by the discriminator. We demonstrate the ability of our active sampling method to achieve higher prediction performance and higher sample efficiency in a domain transfer problem for robotic manipulation tasks. We also present a validation dataset of action-conditioned video of robotic manipulation tasks on which we test the prediction performance of our trained models.
32.Human Grasp Classification for Reactive Human-to-Robot Handovers ⬇️
Transfer of objects between humans and robots is a critical capability for collaborative robots. Although there has been a recent surge of interest in human-robot handovers, most prior research focus on robot-to-human handovers. Further, work on the equally critical human-to-robot handovers often assumes humans can place the object in the robot's gripper. In this paper, we propose an approach for human-to-robot handovers in which the robot meets the human halfway, by classifying the human's grasp of the object and quickly planning a trajectory accordingly to take the object from the human's hand according to their intent. To do this, we collect a human grasp dataset which covers typical ways of holding objects with various hand shapes and poses, and learn a deep model on this dataset to classify the hand grasps into one of these categories. We present a planning and execution approach that takes the object from the human hand according to the detected grasp and hand position, and replans as necessary when the handover is interrupted. Through a systematic evaluation, we demonstrate that our system results in more fluent handovers versus two baselines. We also present findings from a user study (N = 9) demonstrating the effectiveness and usability of our approach with naive users in different scenarios. More results and videos can be found at this http URL.
33.Autoencoders ⬇️
An autoencoder is a specific type of a neural network, which is mainlydesigned to encode the input into a compressed and meaningful representation, andthen decode it back such that the reconstructed input is similar as possible to theoriginal one. This chapter surveys the different types of autoencoders that are mainlyused today. It also describes various applications and use-cases of autoencoders.
34.LiDAR guided Small obstacle Segmentation ⬇️
Detecting small obstacles on the road is critical for autonomous driving. In this paper, we present a method to reliably detect such obstacles through a multi-modal framework of sparse LiDAR(VLP-16) and Monocular vision. LiDAR is employed to provide additional context in the form of confidence maps to monocular segmentation networks. We show significant performance gains when the context is fed as an additional input to monocular semantic segmentation frameworks. We further present a new semantic segmentation dataset to the community, comprising of over 3000 image frames with corresponding LiDAR observations. The images come with pixel-wise annotations of three classes off-road, road, and small obstacle. We stress that precise calibration between LiDAR and camera is crucial for this task and thus propose a novel Hausdorff distance based calibration refinement method over extrinsic parameters. As a first benchmark over this dataset, we report our results with 73% instance detection up to a distance of 50 meters on challenging scenarios. Qualitatively by showcasing accurate segmentation of obstacles less than 15 cms at 50m depth and quantitatively through favourable comparisons vis a vis prior art, we vindicate the method's efficacy. Our project-page and Dataset is hosted at this https URL
35.W2S: A Joint Denoising and Super-Resolution Dataset ⬇️
Denoising and super-resolution (SR) are fundamental tasks in imaging. These two restoration tasks are well covered in the literature, however, only separately. Given a noisy low-resolution (LR) input image, it is yet unclear what the best approach would be in order to obtain a noise-free high-resolution (HR) image. In order to study joint denoising and super-resolution (JDSR), a dataset containing pairs of noisy LR images and the corresponding HR images is fundamental. We propose such a novel JDSR dataset, Wieldfield2SIM (W2S), acquired using microscopy equipment and techniques. W2S is comprised of 144,000 real fluorescence microscopy images, used to form a total of 360 sets of images. A set is comprised of noisy LR images with different noise levels, a noise-free LR image, and a corresponding high-quality HR image. W2S allows us to benchmark the combinations of 6 denoising methods and 6 SR methods. We show that state-of-the-art SR networks perform very poorly on noisy inputs, with a loss reaching 14dB relative to noise-free inputs. Our evaluation also shows that applying the best denoiser in terms of reconstruction error followed by the best SR method does not yield the best result. The best denoising PSNR can, for instance, come at the expense of a loss in high frequencies, which is detrimental for SR methods. We lastly demonstrate that a light-weight SR network with a novel texture loss, trained specifically for JDSR, outperforms any combination of state-of-the-art deep denoising and SR networks.