1.Human Motion Prediction Using Manifold-Aware Wasserstein GAN ⬇️
Human motion prediction aims to forecast future human poses given a prior pose sequence. The discontinuity of the predicted motion and the performance deterioration in long-term horizons are still the main challenges encountered in current literature. In this work, we tackle these issues by using a compact manifold-valued representation of human motion. Specifically, we model the temporal evolution of the 3D human poses as trajectory, what allows us to map human motions to single points on a sphere manifold. To learn these non-Euclidean representations, we build a manifold-aware Wasserstein generative adversarial model that captures the temporal and spatial dependencies of human motion through different losses. Extensive experiments show that our approach outperforms the state-of-the-art on CMU MoCap and Human 3.6M datasets. Our qualitative results show the smoothness of the predicted motions.
2.Content Disentanglement for Semantically Consistent Synthetic-to-RealDomain Adaptation in Urban Traffic Scenes ⬇️
Synthetic data generation is an appealing approach to generate novel traffic scenarios in autonomous driving. However, deep learning techniques trained solely on synthetic data encounter dramatic performance drops when they are tested on real data. Such performance drop is commonly attributed to the domain gap between real and synthetic data. Domain adaptation methods have been applied to mitigate the aforementioned domain gap. These methods achieve visually appealing results, but the translated samples usually introduce semantic inconsistencies. In this work, we propose a new, unsupervised, end-to-end domain adaptation network architecture that enables semantically consistent domain adaptation between synthetic and real data. We evaluate our architecture on the downstream task of semantic segmentation and show that our method achieves superior performance compared to the state-of-the-art methods.
3.Semi-Supervised Classification and Segmentation on High Resolution Aerial Images ⬇️
FloodNet is a high-resolution image dataset acquired by a small UAV platform, DJI Mavic Pro quadcopters, after Hurricane Harvey. The dataset presents a unique challenge of advancing the damage assessment process for post-disaster scenarios using unlabeled and limited labeled dataset. We propose a solution to address their classification and semantic segmentation challenge. We approach this problem by generating pseudo labels for both classification and segmentation during training and slowly incrementing the amount by which the pseudo label loss affects the final loss. Using this semi-supervised method of training helped us improve our baseline supervised loss by a huge margin for classification, allowing the model to generalize and perform better on the validation and test splits of the dataset. In this paper, we compare and contrast the various methods and models for image classification and semantic segmentation on the FloodNet dataset.
4.IntFormer: Predicting pedestrian intention with the aid of the Transformer architecture ⬇️
Understanding pedestrian crossing behavior is an essential goal in intelligent vehicle development, leading to an improvement in their security and traffic flow. In this paper, we developed a method called IntFormer. It is based on transformer architecture and a novel convolutional video classification model called RubiksNet. Following the evaluation procedure in a recent benchmark, we show that our model reaches state-of-the-art results with good performance (
$\approx 40$ seq. per second) and size ($8\times $smaller than the best performing model), making it suitable for real-time usage. We also explore each of the input features, finding that ego-vehicle speed is the most important variable, possibly due to the similarity in crossing cases in PIE dataset.
5.Assessing aesthetics of generated abstract images using correlation structure ⬇️
Can we generate abstract aesthetic images without bias from natural or human selected image corpi? Are aesthetic images singled out in their correlation functions? In this paper we give answers to these and more questions. We generate images using compositional pattern-producing networks with random weights and varying architecture. We demonstrate that even with the randomly selected weights the correlation functions remain largely determined by the network architecture. In a controlled experiment, human subjects picked aesthetic images out of a large dataset of all generated images. Statistical analysis reveals that the correlation function is indeed different for aesthetic images.
6.SAIL-VOS 3D: A Synthetic Dataset and Baselines for Object Detection and 3D Mesh Reconstruction from Video Data ⬇️
Extracting detailed 3D information of objects from video data is an important goal for holistic scene understanding. While recent methods have shown impressive results when reconstructing meshes of objects from a single image, results often remain ambiguous as part of the object is unobserved. Moreover, existing image-based datasets for mesh reconstruction don't permit to study models which integrate temporal information. To alleviate both concerns we present SAIL-VOS 3D: a synthetic video dataset with frame-by-frame mesh annotations which extends SAIL-VOS. We also develop first baselines for reconstruction of 3D meshes from video data via temporal models. We demonstrate efficacy of the proposed baseline on SAIL-VOS 3D and Pix3D, showing that temporal information improves reconstruction quality. Resources and additional information are available at this http URL.
7.ACAE-REMIND for Online Continual Learning with Compressed Feature Replay ⬇️
Online continual learning aims to learn from a non-IID stream of data from a number of different tasks, where the learner is only allowed to consider data once. Methods are typically allowed to use a limited buffer to store some of the images in the stream. Recently, it was found that feature replay, where an intermediate layer representation of the image is stored (or generated) leads to superior results than image replay, while requiring less memory. Quantized exemplars can further reduce the memory usage. However, a drawback of these methods is that they use a fixed (or very intransigent) backbone network. This significantly limits the learning of representations that can discriminate between all tasks. To address this problem, we propose an auxiliary classifier auto-encoder (ACAE) module for feature replay at intermediate layers with high compression rates. The reduced memory footprint per image allows us to save more exemplars for replay. In our experiments, we conduct task-agnostic evaluation under online continual learning setting and get state-of-the-art performance on ImageNet-Subset, CIFAR100 and CIFAR10 dataset.
8.Vision Transformer for Fast and Efficient Scene Text Recognition ⬇️
Scene text recognition (STR) enables computers to read text in natural scenes such as object labels, road signs and instructions. STR helps machines perform informed decisions such as what object to pick, which direction to go, and what is the next step of action. In the body of work on STR, the focus has always been on recognition accuracy. There is little emphasis placed on speed and computational efficiency which are equally important especially for energy-constrained mobile machines. In this paper we propose ViTSTR, an STR with a simple single stage model architecture built on a compute and parameter efficient vision transformer (ViT). On a comparable strong baseline method such as TRBA with accuracy of 84.3%, our small ViTSTR achieves a competitive accuracy of 82.6% (84.2% with data augmentation) at 2.4x speed up, using only 43.4% of the number of parameters and 42.2% FLOPS. The tiny version of ViTSTR achieves 80.3% accuracy (82.1% with data augmentation), at 2.5x the speed, requiring only 10.9% of the number of parameters and 11.9% FLOPS. With data augmentation, our base ViTSTR outperforms TRBA at 85.2% accuracy (83.7% without augmentation) at 2.3x the speed but requires 73.2% more parameters and 61.5% more FLOPS. In terms of trade-offs, nearly all ViTSTR configurations are at or near the frontiers to maximize accuracy, speed and computational efficiency all at the same time.
9.Assessing bikeability with street view imagery and computer vision ⬇️
Studies evaluating bikeability usually compute spatial indicators shaping cycling conditions and conflate them in a quantitative index. Much research involves site visits or conventional geospatial approaches, and few studies have leveraged street view imagery (SVI) for conducting virtual audits. These have assessed a limited range of aspects, and not all have been automated using computer vision (CV). Furthermore, studies have not yet zeroed in on gauging the usability of these technologies thoroughly. We investigate, with experiments at a fine spatial scale and across multiple geographies (Singapore and Tokyo), whether we can use SVI and CV to assess bikeability comprehensively. Extending related work, we develop an exhaustive index of bikeability composed of 34 indicators. The results suggest that SVI and CV are adequate to evaluate bikeability in cities comprehensively. As they outperformed non-SVI counterparts by a wide margin, SVI indicators are also found to be superior in assessing urban bikeability, and potentially can be used independently, replacing traditional techniques. However, the paper exposes some limitations, suggesting that the best way forward is combining both SVI and non-SVI approaches. The new bikeability index presents a contribution in transportation and urban analytics, and it is scalable to assess cycling appeal widely.
10.Unsupervised identification of surgical robotic actions from small non homogeneous datasets ⬇️
Robot-assisted surgery is an established clinical practice. The automatic identification of surgical actions is needed for a range of applications, including performance assessment of trainees and surgical process modeling for autonomous execution and monitoring. However, supervised action identification is not feasible, due to the burden of manually annotating recordings of potentially complex and long surgical executions. Moreover, often few example executions of a surgical procedure can be recorded. This paper proposes a novel algorithm for unsupervised identification of surgical actions in a standard surgical training task, the ring transfer, executed with da Vinci Research Kit. Exploiting kinematic and semantic visual features automatically extracted from a very limited dataset of executions, we are able to significantly outperform the state-of-the-art results for a similar application, improving the quality of segmentation (88% vs. 82% matching score) and clustering (67% vs. 54% F1-score) even in the presence of noise, short actions and non homogeneous workflows, i.e. non repetitive action sequences. Full action identification on hardware with standard commercial specifications is performed in less than 1 s for single execution.
11.Overparametrization of HyperNetworks at Fixed FLOP-Count Enables Fast Neural Image Enhancement ⬇️
Deep convolutional neural networks can enhance images taken with small mobile camera sensors and excel at tasks like demoisaicing, denoising and super-resolution. However, for practical use on mobile devices these networks often require too many FLOPs and reducing the FLOPs of a convolution layer, also reduces its parameter count. This is problematic in view of the recent finding that heavily over-parameterized neural networks are often the ones that generalize best. In this paper we propose to use HyperNetworks to break the fixed ratio of FLOPs to parameters of standard convolutions. This allows us to exceed previous state-of-the-art architectures in SSIM and MS-SSIM on the Zurich RAW- to-DSLR (ZRR) data-set at > 10x reduced FLOP-count. On ZRR we further observe generalization curves consistent with 'double-descent' behavior at fixed FLOP-count, in the large image limit. Finally we demonstrate the same technique can be applied to an existing network (VDN) to reduce its computational cost while maintaining fidelity on the Smartphone Image Denoising Dataset (SIDD). Code for key functions is given in the appendix.
12.Progressively Normalized Self-Attention Network for Video Polyp Segmentation ⬇️
Existing video polyp segmentation (VPS) models typically employ convolutional neural networks (CNNs) to extract features. However, due to their limited receptive fields, CNNs can not fully exploit the global temporal and spatial information in successive video frames, resulting in false-positive segmentation results. In this paper, we propose the novel PNS-Net (Progressively Normalized Self-attention Network), which can efficiently learn representations from polyp videos with real-time speed (~140fps) on a single RTX 2080 GPU and no post-processing. Our PNS-Net is based solely on a basic normalized self-attention block, equipping with recurrence and CNNs entirely. Experiments on challenging VPS datasets demonstrate that the proposed PNS-Net achieves state-of-the-art performance. We also conduct extensive experiments to study the effectiveness of the channel split, soft-attention, and progressive learning strategy. We find that our PNS-Net works well under different settings, making it a promising solution to the VPS task.
13.Unsupervised Compound Domain Adaptation for Face Anti-Spoofing ⬇️
We address the problem of face anti-spoofing which aims to make the face verification systems robust in the real world settings. The context of detecting live vs. spoofed face images may differ significantly in the target domain, when compared to that of labeled source domain where the model is trained. Such difference may be caused due to new and unknown spoof types, illumination conditions, scene backgrounds, among many others. These varieties of differences make the target a compound domain, thus calling for the problem of the unsupervised compound domain adaptation. We demonstrate the effectiveness of the compound domain assumption for the task of face anti-spoofing, for the first time in this work. To this end, we propose a memory augmentation method for adapting the source model to the target domain in a domain aware manner. The adaptation process is further improved by using the curriculum learning and the domain agnostic source network training approaches. The proposed method successfully adapts to the compound target domain consisting multiple new spoof types. Our experiments on multiple benchmark datasets demonstrate the superiority of the proposed method over the state-of-the-art.
14.Deep Active Contours Using Locally Controlled Distance Vector Flow ⬇️
Active contours Model (ACM) has been extensively used in computer vision and image processing. In recent studies, Convolutional Neural Networks (CNNs) have been combined with active contours replacing the user in the process of contour evolution and image segmentation to eliminate limitations associated with ACM's dependence on parameters of the energy functional and initialization. However, prior works did not aim for automatic initialization which is addressed here. In addition to manual initialization, current methods are highly sensitive to initial location and fail to delineate borders accurately. We propose a fully automatic image segmentation method to address problems of manual initialization, insufficient capture range, and poor convergence to boundaries, in addition to the problem of assignment of energy functional parameters. We train two CNNs, which predict active contour weighting parameters and generate a ground truth mask to extract Distance Transform (DT) and an initialization circle. Distance transform is used to form a vector field pointing from each pixel of the image towards the closest point on the boundary, the size of which is equal to the Euclidean distance map. We evaluate our method on four publicly available datasets including two building instance segmentation datasets, Vaihingen and Bing huts, and two mammography image datasets, INBreast and DDSM-BCRP. Our approach outperforms latest research by 0.59 ans 2.39 percent in mean Intersection-over-Union (mIoU), 7.38 and 8.62 percent in Boundary F-score (BoundF) for Vaihingen and Bing huts datasets, respectively. Dice similarity coefficient for the INBreast and DDSM-BCRP datasets is 94.23% and 90.89%, respectively indicating our method is comparable to state-of-the-art frameworks.
15.Improved detection of small objects in road network sequences ⬇️
The vast number of existing IP cameras in current road networks is an opportunity to take advantage of the captured data and analyze the video and detect any significant events. For this purpose, it is necessary to detect moving vehicles, a task that was carried out using classical artificial vision techniques until a few years ago. Nowadays, significant improvements have been obtained by deep learning networks. Still, object detection is considered one of the leading open issues within computer vision.
The current scenario is constantly evolving, and new models and techniques are appearing trying to improve this field. In particular, new problems and drawbacks appear regarding detecting small objects, which correspond mainly to the vehicles that appear in the road scenes. All this means that new solutions that try to improve the low detection rate of small elements are essential. Among the different emerging research lines, this work focuses on the detection of small objects. In particular, our proposal aims to vehicle detection from images captured by video surveillance cameras.
In this work, we propose a new procedure for detecting small-scale objects by applying super-resolution processes based on detections performed by convolutional neural networks \emph{(CNN)}. The neural network is integrated with processes that are in charge of increasing the resolution of the images to improve the object detection performance. This solution has been tested for a set of traffic images containing elements of different scales to test the efficiency according to the detections obtained by the model, thus demonstrating that our proposal achieves good results in a wide range of situations.
16.I2C2W: Image-to-Character-to-Word Transformers for Accurate Scene Text Recognition ⬇️
Leveraging the advances of natural language processing, most recent scene text recognizers adopt an encoder-decoder architecture where text images are first converted to representative features and then a sequence of characters via `direct decoding'. However, scene text images suffer from rich noises of different sources such as complex background and geometric distortions which often confuse the decoder and lead to incorrect alignment of visual features at noisy decoding time steps. This paper presents I2C2W, a novel scene text recognizer that is accurate and tolerant to various noises in scenes. I2C2W consists of an image-to-character module (I2C) and a character-to-word module (C2W) which are complementary and can be trained end-to-end. I2C detects characters and predicts their relative positions in a word. It strives to detect all possible characters including incorrect and redundant ones based on different alignments of visual features without the restriction of time steps. Taking the detected characters as input, C2W learns from character semantics and their positions to filter out incorrect and redundant detection and produce the final word recognition. Extensive experiments over seven public datasets show that I2C2W achieves superior recognition performances and outperforms the state-of-the-art by large margins on challenging irregular scene text datasets.
17.Exemplar-Based Open-Set Panoptic Segmentation Network ⬇️
We extend panoptic segmentation to the open-world and introduce an open-set panoptic segmentation~(OPS) task. This task requires performing panoptic segmentation for not only \known classes but also \unknown ones that have not been acknowledged during training. We investigate the practical challenges of the task and construct a benchmark on top of an existing dataset, COCO. In addition, we propose a novel exemplar-based open-set panoptic segmentation network~(EOPSN) inspired by exemplar theory. Our approach identifies a new class based on exemplars, which are identified by clustering and employed as pseudo-ground-truths. The size of each class increases by mining new exemplars based on the similarities to the existing ones associated with the class. We evaluate EOPSN on the proposed benchmark and demonstrate the effectiveness of our proposals. The primary goal of our work is to draw the attention of the community to the recognition in the open-world scenarios. The implementation of our algorithm is available on the project webpage: this https URL.
18.Exploring Driving-aware Salient Object Detection via Knowledge Transfer ⬇️
Recently, general salient object detection (SOD) has made great progress with the rapid development of deep neural networks. However, task-aware SOD has hardly been studied due to the lack of task-specific datasets. In this paper, we construct a driving task-oriented dataset where pixel-level masks of salient objects have been annotated. Comparing with general SOD datasets, we find that the cross-domain knowledge difference and task-specific scene gap are two main challenges to focus the salient objects when driving. Inspired by these findings, we proposed a baseline model for the driving task-aware SOD via a knowledge transfer convolutional neural network. In this network, we construct an attentionbased knowledge transfer module to make up the knowledge difference. In addition, an efficient boundary-aware feature decoding module is introduced to perform fine feature decoding for objects in the complex task-specific scenes. The whole network integrates the knowledge transfer and feature decoding modules in a progressive manner. Experiments show that the proposed dataset is very challenging, and the proposed method outperforms 12 state-of-the-art methods on the dataset, which facilitates the development of task-aware SOD.
19.NExT-QA:Next Phase of Question-Answering to Explaining Temporal Actions ⬇️
We introduce NExT-QA, a rigorously designed video question answering (VideoQA) benchmark to advance video understanding from describing to explaining the temporal actions. Based on the dataset, we set up multi-choice and open-ended QA tasks targeting causal action reasoning, temporal action reasoning, and common scene comprehension. Through extensive analysis of baselines and established VideoQA techniques, we find that top-performing methods excel at shallow scene descriptions but are weak in causal and temporal action reasoning. Furthermore, the models that are effective on multi-choice QA, when adapted to open-ended QA, still struggle in generalizing the answers. This raises doubt on the ability of these models to reason and highlights possibilities for improvement. With detailed results for different question types and heuristic observations for future works, we hope NExT-QA will guide the next generation of VQA research to go beyond superficial scene description towards a deeper understanding of videos. (The dataset and related resources are available at this https URL)
20.Finding a Needle in a Haystack: Tiny Flying Object Detection in 4K Videos using a Joint Detection-and-Tracking Approach ⬇️
Detecting tiny objects in a high-resolution video is challenging because the visual information is little and unreliable. Specifically, the challenge includes very low resolution of the objects, MPEG artifacts due to compression and a large searching area with many hard negatives. Tracking is equally difficult because of the unreliable appearance, and the unreliable motion estimation. Luckily, we found that by combining this two challenging tasks together, there will be mutual benefits. Following the idea, in this paper, we present a neural network model called the Recurrent Correlational Network, where detection and tracking are jointly performed over a multi-frame representation learned through a single, trainable, and end-to-end network. The framework exploits a convolutional long short-term memory network for learning informative appearance changes for detection, while the learned representation is shared in tracking for enhancing its performance. In experiments with datasets containing images of scenes with small flying objects, such as birds and unmanned aerial vehicles, the proposed method yielded consistent improvements in detection performance over deep single-frame detectors and existing motion-based detectors. Furthermore, our network performs as well as state-of-the-art generic object trackers when it was evaluated as a tracker on a bird image dataset.
21.Weakly Supervised Dense Video Captioning via Jointly Usage of Knowledge Distillation and Cross-modal Matching ⬇️
This paper proposes an approach to Dense Video Captioning (DVC) without pairwise event-sentence annotation. First, we adopt the knowledge distilled from relevant and well solved tasks to generate high-quality event proposals. Then we incorporate contrastive loss and cycle-consistency loss typically applied to cross-modal retrieval tasks to build semantic matching between the proposals and sentences, which are eventually used to train the caption generation module. In addition, the parameters of matching module are initialized via pre-training based on annotated images to improve the matching performance. Extensive experiments on ActivityNet-Caption dataset reveal the significance of distillation-based event proposal generation and cross-modal retrieval-based semantic matching to weakly supervised DVC, and demonstrate the superiority of our method to existing state-of-the-art methods.
22.Self-Point-Flow: Self-Supervised Scene Flow Estimation from Point Clouds with Optimal Transport and Random Walk ⬇️
Due to the scarcity of annotated scene flow data, self-supervised scene flow learning in point clouds has attracted increasing attention. In the self-supervised manner, establishing correspondences between two point clouds to approximate scene flow is an effective approach. Previous methods often obtain correspondences by applying point-wise matching that only takes the distance on 3D point coordinates into account, introducing two critical issues: (1) it overlooks other discriminative measures, such as color and surface normal, which often bring fruitful clues for accurate matching; and (2) it often generates sub-par performance, as the matching is operated in an unconstrained situation, where multiple points can be ended up with the same corresponding point. To address the issues, we formulate this matching task as an optimal transport problem. The output optimal assignment matrix can be utilized to guide the generation of pseudo ground truth. In this optimal transport, we design the transport cost by considering multiple descriptors and encourage one-to-one matching by mass equality constraints. Also, constructing a graph on the points, a random walk module is introduced to encourage the local consistency of the pseudo labels. Comprehensive experiments on FlyingThings3D and KITTI show that our method achieves state-of-the-art performance among self-supervised learning methods. Our self-supervised method even performs on par with some supervised learning approaches, although we do not need any ground truth flow for training.
23.Towards Unsupervised Sketch-based Image Retrieval ⬇️
Current supervised sketch-based image retrieval (SBIR) methods achieve excellent performance. However, the cost of data collection and labeling imposes an intractable barrier to practical deployment of real applications. In this paper, we present the first attempt at unsupervised SBIR to remove the labeling cost (category annotations and sketch-photo pairings) that is conventionally needed for training. Existing single-domain unsupervised representation learning methods perform poorly in this application, due to the unique cross-domain (sketch and photo) nature of the problem. We therefore introduce a novel framework that simultaneously performs unsupervised representation learning and sketch-photo domain alignment. Technically this is underpinned by exploiting joint distribution optimal transport (JDOT) to align data from different domains during representation learning, which we extend with trainable cluster prototypes and feature memory banks to further improve scalability and efficacy. Extensive experiments show that our framework achieves excellent performance in the new unsupervised setting, and performs comparably or better than state-of-the-art in the zero-shot setting.
24.Single View Geocentric Pose in the Wild ⬇️
Current methods for Earth observation tasks such as semantic mapping, map alignment, and change detection rely on near-nadir images; however, often the first available images in response to dynamic world events such as natural disasters are oblique. These tasks are much more difficult for oblique images due to observed object parallax. There has been recent success in learning to regress geocentric pose, defined as height above ground and orientation with respect to gravity, by training with airborne lidar registered to satellite images. We present a model for this novel task that exploits affine invariance properties to outperform state of the art performance by a wide margin. We also address practical issues required to deploy this method in the wild for real-world applications. Our data and code are publicly available.
25.Decorating Your Own Bedroom: Locally Controlling Image Generation with Generative Adversarial Networks ⬇️
Generative Adversarial Networks (GANs) have made great success in synthesizing high-quality images. However, how to steer the generation process of a well-trained GAN model and customize the output image is much less explored. It has been recently found that modulating the input latent code used in GANs can reasonably alter some variation factors in the output image, but such manipulation usually presents to change the entire image as a whole. In this work, we propose an effective approach, termed as LoGAN, to support local editing of the output image. Concretely, we introduce two operators, i.e., content modulation and style modulation, together with a priority mask to facilitate the precise control of the intermediate generative features. Taking bedroom synthesis as an instance, we are able to seamlessly remove, insert, shift, and rotate the individual objects inside a room. Furthermore, our method can completely clear out a room and then refurnish it with customized furniture and styles. Experimental results show the great potentials of steering the image generation of pre-trained GANs for versatile image editing.
26.Reinforcement Learning for Adaptive Video Compressive Sensing ⬇️
We apply reinforcement learning to video compressive sensing to adapt the compression ratio. Specifically, video snapshot compressive imaging (SCI), which captures high-speed video using a low-speed camera is considered in this work, in which multiple (B) video frames can be reconstructed from a snapshot measurement. One research gap in previous studies is how to adapt B in the video SCI system for different scenes. In this paper, we fill this gap utilizing reinforcement learning (RL). An RL model, as well as various convolutional neural networks for reconstruction, are learned to achieve adaptive sensing of video SCI systems. Furthermore, the performance of an object detection network using directly the video SCI measurements without reconstruction is also used to perform RL-based adaptive video compressive sensing. Our proposed adaptive SCI method can thus be implemented in low cost and real time. Our work takes the technology one step further towards real applications of video SCI.
27.Physically Plausible Pose Refinement using Fully Differentiable Forces ⬇️
All hand-object interaction is controlled by forces that the two bodies exert on each other, but little work has been done in modeling these underlying forces when doing pose and contact estimation from RGB/RGB-D data. Given the pose of the hand and object from any pose estimation system, we propose an end-to-end differentiable model that refines pose estimates by learning the forces experienced by the object at each vertex in its mesh. By matching the learned net force to an estimate of net force based on finite differences of position, this model is able to find forces that accurately describe the movement of the object, while resolving issues like mesh interpenetration and lack of contact. Evaluating on the ContactPose dataset, we show this model successfully corrects poses and finds contact maps that better match the ground truth, despite not using any RGB or depth image data.
28.Visual FUDGE: Form Understanding via Dynamic Graph Editing ⬇️
We address the problem of form understanding: finding text entities and the relationships/links between them in form images. The proposed FUDGE model formulates this problem on a graph of text elements (the vertices) and uses a Graph Convolutional Network to predict changes to the graph. The initial vertices are detected text lines and do not necessarily correspond to the final text entities, which can span multiple lines. Also, initial edges contain many false-positive relationships. FUDGE edits the graph structure by combining text segments (graph vertices) and pruning edges in an iterative fashion to obtain the final text entities and relationships. While recent work in this area has focused on leveraging large-scale pre-trained Language Models (LM), FUDGE achieves the same level of entity linking performance on the FUNSD dataset by learning only visual features from the (small) provided training set. FUDGE can be applied on forms where text recognition is difficult (e.g. degraded or historical forms) and on forms in resource-poor languages where pre-training such LMs is challenging. FUDGE is state-of-the-art on the historical NAF dataset.
29.Graph Neural Networks for Knowledge Enhanced Visual Representation of Paintings ⬇️
We propose ArtSAGENet, a novel multimodal architecture that integrates Graph Neural Networks (GNNs) and Convolutional Neural Networks (CNNs), to jointly learn visual and semantic-based artistic representations. First, we illustrate the significant advantages of multi-task learning for fine art analysis and argue that it is conceptually a much more appropriate setting in the fine art domain than the single-task alternatives. We further demonstrate that several GNN architectures can outperform strong CNN baselines in a range of fine art analysis tasks, such as style classification, artist attribution, creation period estimation, and tag prediction, while training them requires an order of magnitude less computational time and only a small amount of labeled data. Finally, through extensive experimentation we show that our proposed ArtSAGENet captures and encodes valuable relational dependencies between the artists and the artworks, surpassing the performance of traditional methods that rely solely on the analysis of visual content. Our findings underline a great potential of integrating visual content and semantics for fine art analysis and curation.
30.Deep Metric Learning for Few-Shot Image Classification: A Selective Review ⬇️
Few-shot image classification is a challenging problem which aims to achieve the human level of recognition based only on a small number of images. Deep learning algorithms such as meta-learning, transfer learning, and metric learning have been employed recently and achieved the state-of-the-art performance. In this survey, we review representative deep metric learning methods for few-shot classification, and categorize them into three groups according to the major problems and novelties they focus on. We conclude this review with a discussion on current challenges and future trends in few-shot image classification.
31.VPN++: Rethinking Video-Pose embeddings for understanding Activities of Daily Living ⬇️
Many attempts have been made towards combining RGB and 3D poses for the recognition of Activities of Daily Living (ADL). ADL may look very similar and often necessitate to model fine-grained details to distinguish them. Because the recent 3D ConvNets are too rigid to capture the subtle visual patterns across an action, this research direction is dominated by methods combining RGB and 3D Poses. But the cost of computing 3D poses from RGB stream is high in the absence of appropriate sensors. This limits the usage of aforementioned approaches in real-world applications requiring low latency. Then, how to best take advantage of 3D Poses for recognizing ADL? To this end, we propose an extension of a pose driven attention mechanism: Video-Pose Network (VPN), exploring two distinct directions. One is to transfer the Pose knowledge into RGB through a feature-level distillation and the other towards mimicking pose driven attention through an attention-level distillation. Finally, these two approaches are integrated into a single model, we call VPN++. We show that VPN++ is not only effective but also provides a high speed up and high resilience to noisy Poses. VPN++, with or without 3D Poses, outperforms the representative baselines on 4 public datasets. Code is available at this https URL.
32.PixMatch: Unsupervised Domain Adaptation via Pixelwise Consistency Training ⬇️
Unsupervised domain adaptation is a promising technique for semantic segmentation and other computer vision tasks for which large-scale data annotation is costly and time-consuming. In semantic segmentation, it is attractive to train models on annotated images from a simulated (source) domain and deploy them on real (target) domains. In this work, we present a novel framework for unsupervised domain adaptation based on the notion of target-domain consistency training. Intuitively, our work is based on the idea that in order to perform well on the target domain, a model's output should be consistent with respect to small perturbations of inputs in the target domain. Specifically, we introduce a new loss term to enforce pixelwise consistency between the model's predictions on a target image and a perturbed version of the same image. In comparison to popular adversarial adaptation methods, our approach is simpler, easier to implement, and more memory-efficient during training. Experiments and extensive ablation studies demonstrate that our simple approach achieves remarkably strong results on two challenging synthetic-to-real benchmarks, GTA5-to-Cityscapes and SYNTHIA-to-Cityscapes.
Code is available at: this https URL
33.Finding an Unsupervised Image Segmenter in Each of Your Deep Generative Models ⬇️
Recent research has shown that numerous human-interpretable directions exist in the latent space of GANs. In this paper, we develop an automatic procedure for finding directions that lead to foreground-background image separation, and we use these directions to train an image segmentation model without human supervision. Our method is generator-agnostic, producing strong segmentation results with a wide range of different GAN architectures. Furthermore, by leveraging GANs pretrained on large datasets such as ImageNet, we are able to segment images from a range of domains without further training or finetuning. Evaluating our method on image segmentation benchmarks, we compare favorably to prior work while using neither human supervision nor access to the training data. Broadly, our results demonstrate that automatically extracting foreground-background structure from pretrained deep generative models can serve as a remarkably effective substitute for human supervision.
34.Fighting Gradients with Gradients: Dynamic Defenses against Adversarial Attacks ⬇️
Adversarial attacks optimize against models to defeat defenses. Existing defenses are static, and stay the same once trained, even while attacks change. We argue that models should fight back, and optimize their defenses against attacks at test time. We propose dynamic defenses, to adapt the model and input during testing, by defensive entropy minimization (dent). Dent alters testing, but not training, for compatibility with existing models and train-time defenses. Dent improves the robustness of adversarially-trained defenses and nominally-trained models against white-box, black-box, and adaptive attacks on CIFAR-10/100 and ImageNet. In particular, dent boosts state-of-the-art defenses by 20+ points absolute against AutoAttack on CIFAR-10 at
$\epsilon_\infty$ = 8/255.
35.Image Cropping on Twitter: Fairness Metrics, their Limitations, and the Importance of Representation, Design, and Agency ⬇️
Twitter uses machine learning to crop images, where crops are centered around the part predicted to be the most salient. In fall 2020, Twitter users raised concerns that the automated image cropping system on Twitter favored light-skinned over dark-skinned individuals, as well as concerns that the system favored cropping woman's bodies instead of their heads. In order to address these concerns, we conduct an extensive analysis using formalized group fairness metrics. We find systematic disparities in cropping and identify contributing factors, including the fact that the cropping based on the single most salient point can amplify the disparities. However, we demonstrate that formalized fairness metrics and quantitative analysis on their own are insufficient for capturing the risk of representational harm in automatic cropping. We suggest the removal of saliency-based cropping in favor of a solution that better preserves user agency. For developing a new solution that sufficiently address concerns related to representational harm, our critique motivates a combination of quantitative and qualitative methods that include human-centered design.
36.A multimodal deep learning framework for scalable content based visual media retrieval ⬇️
We propose a novel, efficient, modular and scalable framework for content based visual media retrieval systems by leveraging the power of Deep Learning which is flexible to work both for images and videos conjointly and we also introduce an efficient comparison and filtering metric for retrieval. We put forward our findings from critical performance tests comparing our method to the predominant conventional approach to demonstrate the feasibility and efficiency of the proposed solution with best practices, possible improvements that may further augment the ability of retrieval architectures.
37.Fast and Accurate Single-Image Depth Estimation on Mobile Devices, Mobile AI 2021 Challenge: Report ⬇️
Depth estimation is an important computer vision problem with many practical applications to mobile devices. While many solutions have been proposed for this task, they are usually very computationally expensive and thus are not applicable for on-device inference. To address this problem, we introduce the first Mobile AI challenge, where the target is to develop an end-to-end deep learning-based depth estimation solutions that can demonstrate a nearly real-time performance on smartphones and IoT platforms. For this, the participants were provided with a new large-scale dataset containing RGB-depth image pairs obtained with a dedicated stereo ZED camera producing high-resolution depth maps for objects located at up to 50 meters. The runtime of all models was evaluated on the popular Raspberry Pi 4 platform with a mobile ARM-based Broadcom chipset. The proposed solutions can generate VGA resolution depth maps at up to 10 FPS on the Raspberry Pi 4 while achieving high fidelity results, and are compatible with any Android or Linux-based mobile devices. A detailed description of all models developed in the challenge is provided in this paper.
38.Fast Camera Image Denoising on Mobile GPUs with Deep Learning, Mobile AI 2021 Challenge: Report ⬇️
Image denoising is one of the most critical problems in mobile photo processing. While many solutions have been proposed for this task, they are usually working with synthetic data and are too computationally expensive to run on mobile devices. To address this problem, we introduce the first Mobile AI challenge, where the target is to develop an end-to-end deep learning-based image denoising solution that can demonstrate high efficiency on smartphone GPUs. For this, the participants were provided with a novel large-scale dataset consisting of noisy-clean image pairs captured in the wild. The runtime of all models was evaluated on the Samsung Exynos 2100 chipset with a powerful Mali GPU capable of accelerating floating-point and quantized neural networks. The proposed solutions are fully compatible with any mobile GPU and are capable of processing 480p resolution images under 40-80 ms while achieving high fidelity results. A detailed description of all models developed in the challenge is provided in this paper.
39.Detecting Adversarial Examples with Bayesian Neural Network ⬇️
Deep neural networks (DNNs) are vulnerable against adversarial examples, i.e., examples that are carefully crafted to fool the DNNs while being indistinguishable from the natural images to humans. In this paper, we propose a new framework to detect adversarial examples motivated by the observations that random components can improve the smoothness of predictors and make it easier to simulate output distribution of deep neural network. With these observations, we propose a novel Bayesian adversarial example detector, short for BATector, to improve the performance of adversarial example detection. In specific, we study the distributional difference of hidden layer output between natural and adversarial examples, and propose to use the randomness of Bayesian neural network (BNN) to simulate hidden layer output distribution and leverage the distribution dispersion to detect adversarial examples. The advantage of BNN is that the output is stochastic while neural network without random components do not have such characteristics. Empirical results on several benchmark datasets against popular attacks show that, the proposed BATector outperforms the state-of-the-art detectors in adversarial example detection.
40.Shape Analysis of Functional Data with Elastic Partial Matching ⬇️
Elastic Riemannian metrics have been used successfully in the past for statistical treatments of functional and curve shape data. However, this usage has suffered from an important restriction: the function boundaries are assumed fixed and matched. Functional data exhibiting unmatched boundaries typically arise from dynamical systems with variable evolution rates such as COVID-19 infection rate curves associated with different geographical regions. In this case, it is more natural to model such data with sliding boundaries and use partial matching, i.e., only a part of a function is matched to another function. Here, we develop a comprehensive Riemannian framework that allows for partial matching, comparing, and clustering of functions under both phase variability and uncertain boundaries. We extend past work by: (1) Forming a joint action of the time-warping and time-scaling groups; (2) Introducing a metric that is invariant to this joint action, allowing for a gradient-based approach to elastic partial matching; and (3) Presenting a modification that, while losing the metric property, allows one to control relative influence of the two groups. This framework is illustrated for registering and clustering shapes of COVID-19 rate curves, identifying essential patterns, minimizing mismatch errors, and reducing variability within clusters compared to previous methods.
41.UncertaintyFuseNet: Robust Uncertainty-aware Hierarchical Feature Fusion with Ensemble Monte Carlo Dropout for COVID-19 Detection ⬇️
The COVID-19 (Coronavirus disease 2019) has infected more than 151 million people and caused approximately 3.17 million deaths around the world up to the present. The rapid spread of COVID-19 is continuing to threaten human's life and health. Therefore, the development of computer-aided detection (CAD) systems based on machine and deep learning methods which are able to accurately differentiate COVID-19 from other diseases using chest computed tomography (CT) and X-Ray datasets is essential and of immediate priority. Different from most of the previous studies which used either one of CT or X-ray images, we employed both data types with sufficient samples in implementation. On the other hand, due to the extreme sensitivity of this pervasive virus, model uncertainty should be considered, while most previous studies have overlooked it. Therefore, we propose a novel powerful fusion model named
$UncertaintyFuseNet$ that consists of an uncertainty module: Ensemble Monte Carlo (EMC) dropout. The obtained results prove the effectiveness of our proposed fusion for COVID-19 detection using CT scan and X-Ray datasets. Also, our proposed$UncertaintyFuseNet$ model is significantly robust to noise and performs well with the previously unseen data. The source codes and models of this study are available at: this https URL.
42.Contrastive Model Inversion for Data-Free Knowledge Distillation ⬇️
Model inversion, whose goal is to recover training data from a pre-trained model, has been recently proved feasible. However, existing inversion methods usually suffer from the mode collapse problem, where the synthesized instances are highly similar to each other and thus show limited effectiveness for downstream tasks, such as knowledge distillation. In this paper, we propose Contrastive Model Inversion~(CMI), where the data diversity is explicitly modeled as an optimizable objective, to alleviate the mode collapse issue. Our main observation is that, under the constraint of the same amount of data, higher data diversity usually indicates stronger instance discrimination. To this end, we introduce in CMI a contrastive learning objective that encourages the synthesizing instances to be distinguishable from the already synthesized ones in previous batches. Experiments of pre-trained models on CIFAR-10, CIFAR-100, and Tiny-ImageNet demonstrate that CMI not only generates more visually plausible instances than the state of the arts, but also achieves significantly superior performance when the generated data are used for knowledge distillation. Code is available at \url{this https URL}.
43.Dependent Multi-Task Learning with Causal Intervention for Image Captioning ⬇️
Recent work for image captioning mainly followed an extract-then-generate paradigm, pre-extracting a sequence of object-based features and then formulating image captioning as a single sequence-to-sequence task. Although promising, we observed two problems in generated captions: 1) content inconsistency where models would generate contradicting facts; 2) not informative enough where models would miss parts of important information. From a causal perspective, the reason is that models have captured spurious statistical correlations between visual features and certain expressions (e.g., visual features of "long hair" and "woman"). In this paper, we propose a dependent multi-task learning framework with the causal intervention (DMTCI). Firstly, we involve an intermediate task, bag-of-categories generation, before the final task, image captioning. The intermediate task would help the model better understand the visual features and thus alleviate the content inconsistency problem. Secondly, we apply Pearl's do-calculus on the model, cutting off the link between the visual features and possible confounders and thus letting models focus on the causal visual features. Specifically, the high-frequency concept set is considered as the proxy confounders where the real confounders are inferred in the continuous space. Finally, we use a multi-agent reinforcement learning (MARL) strategy to enable end-to-end training and reduce the inter-task error accumulations. The extensive experiments show that our model outperforms the baseline models and achieves competitive performance with state-of-the-art models.
44.Fixed $β$ -VAE Encoding for Curious Exploration in Complex 3D Environments ⬇️
Curiosity is a general method for augmenting an environment reward with an intrinsic reward, which encourages exploration and is especially useful in sparse reward settings. As curiosity is calculated using next state prediction error, the type of state encoding used has a large impact on performance. Random features and inverse-dynamics features are generally preferred over VAEs based on previous results from Atari and other mostly 2D environments. However, unlike VAEs, they may not encode sufficient information for optimal behaviour, which becomes increasingly important as environments become more complex. In this paper, we use the sparse reward 3D physics environment Animal-AI, to demonstrate how a fixed
$\beta$ -VAE encoding can be used effectively with curiosity. We combine this with curriculum learning to solve the previously unsolved exploration intensive detour tasks while achieving 22% gain in sample efficiency on the training curriculum against the next best encoding. We also corroborate the results on Atari Breakout, with our custom encoding outperforming random features and inverse-dynamics features.
45.COVID-19 Detection in Computed Tomography Images with 2D and 3D Approaches ⬇️
Detecting COVID-19 in computed tomography (CT) or radiography images has been proposed as a supplement to the definitive RT-PCR test. We present a deep learning ensemble for detecting COVID-19 infection, combining slice-based (2D) and volume-based (3D) approaches. The 2D system detects the infection on each CT slice independently, combining them to obtain the patient-level decision via different methods (averaging and long-short term memory networks). The 3D system takes the whole CT volume to arrive to the patient-level decision in one step. A new high resolution chest CT scan dataset, called the IST-C dataset, is also collected in this work. The proposed ensemble, called IST-CovNet, obtains 90.80% accuracy and 0.95 AUC score overall on the IST-C dataset in detecting COVID-19 among normal controls and other types of lung pathologies; and 93.69% accuracy and 0.99 AUC score on the publicly available MosMed dataset that consists of COVID-19 scans and normal controls only. The system is deployed at Istanbul University Cerrahpasa School of Medicine.
46.Multi-view Contrastive Coding of Remote Sensing Images at Pixel-level ⬇️
Our planet is viewed by satellites through multiple sensors (e.g., multi-spectral, Lidar and SAR) and at different times. Multi-view observations bring us complementary information than the single one. Alternatively, there are common features shared between different views, such as geometry and semantics. Recently, contrastive learning methods have been proposed for the alignment of multi-view remote sensing images and improving the feature representation of single sensor images by modeling view-invariant factors. However, these methods are based on the pretraining of the predefined tasks or just focus on image-level classification. Moreover, these methods lack research on uncertainty estimation. In this work, a pixel-wise contrastive approach based on an unlabeled multi-view setting is proposed to overcome this limitation. This is achieved by the use of contrastive loss in the feature alignment and uniformity between multi-view images. In this approach, a pseudo-Siamese ResUnet is trained to learn a representation that aims to align features from the shifted positive pairs and uniform the induced distribution of the features on the hypersphere. The learned features of multi-view remote sensing images are evaluated on a liner protocol evaluation and an unsupervised change detection task. We analyze key properties of the approach that make it work, finding that the requirement of shift equivariance ensured the success of the proposed approach and the uncertainty estimation of representations leads to performance improvements. Moreover, the performance of multi-view contrastive learning is affected by the choice of different sensors. Results demonstrate both improvements in efficiency and accuracy over the state-of-the-art multi-view contrastive methods.
47.Parallel Attention Network with Sequence Matching for Video Grounding ⬇️
Given a video, video grounding aims to retrieve a temporal moment that semantically corresponds to a language query. In this work, we propose a Parallel Attention Network with Sequence matching (SeqPAN) to address the challenges in this task: multi-modal representation learning, and target moment boundary prediction. We design a self-guided parallel attention module to effectively capture self-modal contexts and cross-modal attentive information between video and text. Inspired by sequence labeling tasks in natural language processing, we split the ground truth moment into begin, inside, and end regions. We then propose a sequence matching strategy to guide start/end boundary predictions using region labels. Experimental results on three datasets show that SeqPAN is superior to state-of-the-art methods. Furthermore, the effectiveness of the self-guided parallel attention module and the sequence matching module is verified.
48.Transfer learning approach to Classify the X-ray image that corresponds to corona disease Using ResNet50 pretrained by ChexNet ⬇️
Coronavirus adversely has affected people worldwide. There are common symptoms between the Covid19 virus disease and other respiratory diseases like pneumonia or Influenza. Therefore, diagnosing it fast is crucial not only to save patients but also to prevent it from spreading. One of the most reliant methods of diagnosis is through X-ray images of a lung. With the help of deep learning approaches, we can teach the deep model to learn the condition of an affected lung. Therefore, it can classify the new sample as if it is a Covid19 infected patient or not. In this project, we train a deep model based on ResNet50 pretrained by ImageNet dataset and CheXNet dataset. Based on the imbalanced CoronaHack Chest X-Ray dataset introducing by Kaggle we applied both binary and multi-class classification. Also, we compare the results when using Focal loss and Cross entropy loss.
49.Sparta: Spatially Attentive and Adversarially Robust Activation ⬇️
Adversarial training (AT) is one of the most effective ways for improving the robustness of deep convolution neural networks (CNNs). Just like common network training, the effectiveness of AT relies on the design of basic network components. In this paper, we conduct an in-depth study on the role of the basic ReLU activation component in AT for robust CNNs. We find that the spatially-shared and input-independent properties of ReLU activation make CNNs less robust to white-box adversarial attacks with either standard or adversarial training. To address this problem, we extend ReLU to a novel Sparta activation function (Spatially attentive and Adversarially Robust Activation), which enables CNNs to achieve both higher robustness, i.e., lower error rate on adversarial examples, and higher accuracy, i.e., lower error rate on clean examples, than the existing state-of-the-art (SOTA) activation functions. We further study the relationship between Sparta and the SOTA activation functions, providing more insights about the advantages of our method. With comprehensive experiments, we also find that the proposed method exhibits superior cross-CNN and cross-dataset transferability. For the former, the adversarially trained Sparta function for one CNN (e.g., ResNet-18) can be fixed and directly used to train another adversarially robust CNN (e.g., ResNet-34). For the latter, the Sparta function trained on one dataset (e.g., CIFAR-10) can be employed to train adversarially robust CNNs on another dataset (e.g., SVHN). In both cases, Sparta leads to CNNs with higher robustness than the vanilla ReLU, verifying the flexibility and versatility of the proposed method.
50.EchoCP: An Echocardiography Dataset in Contrast Transthoracic Echocardiography for Patent Foramen Ovale Diagnosis ⬇️
Patent foramen ovale (PFO) is a potential separation between the septum, primum and septum secundum located in the anterosuperior portion of the atrial septum. PFO is one of the main factors causing cryptogenic stroke which is the fifth leading cause of death in the United States. For PFO diagnosis, contrast transthoracic echocardiography (cTTE) is preferred as being a more robust method compared with others. However, the current PFO diagnosis through cTTE is extremely slow as it is proceeded manually by sonographers on echocardiography videos. Currently there is no publicly available dataset for this important topic in the community. In this paper, we present EchoCP, as the first echocardiography dataset in cTTE targeting PFO diagnosis.
EchoCP consists of 30 patients with both rest and Valsalva maneuver videos which covers various PFO grades. We further establish an automated baseline method for PFO diagnosis based on the state-of-the-art cardiac chamber segmentation technique, which achieves 0.89 average mean Dice score, but only 0.70/0.67 mean accuracies for PFO diagnosis, leaving large room for improvement. We hope that the challenging EchoCP dataset can stimulate further research and lead to innovative and generic solutions that would have an impact in multiple domains. Our dataset is released.
51.Randomly Initialized Convolutional Neural Network for the Recognition of COVID-19 using X-ray Images ⬇️
By the start of 2020, the novel coronavirus disease (COVID-19) has been declared a worldwide pandemic. Because of the severity of this infectious disease, several kinds of research have focused on combatting its ongoing spread. One potential solution to detect COVID-19 is by analyzing the chest X-ray images using Deep Learning (DL) models. In this context, Convolutional Neural Networks (CNNs) are presented as efficient techniques for early diagnosis. In this study, we propose a novel randomly initialized CNN architecture for the recognition of COVID-19. This network consists of a set of different-sized hidden layers created from scratch. The performance of this network is evaluated through two public datasets, which are the COVIDx and the enhanced COVID-19 datasets. Both of these datasets consist of 3 different classes of images: COVID19, pneumonia, and normal chest X-ray images. The proposed CNN model yields encouraging results with 94% and 99% of accuracy for COVIDx and enhanced COVID-19 dataset, respectively.
52.Transfer Learning Enhanced Generative Adversarial Networks for Multi-Channel MRI Reconstruction ⬇️
Deep learning based generative adversarial networks (GAN) can effectively perform image reconstruction with under-sampled MR data. In general, a large number of training samples are required to improve the reconstruction performance of a certain model. However, in real clinical applications, it is difficult to obtain tens of thousands of raw patient data to train the model since saving k-space data is not in the routine clinical flow. Therefore, enhancing the generalizability of a network based on small samples is urgently needed. In this study, three novel applications were explored based on parallel imaging combined with the GAN model (PI-GAN) and transfer learning. The model was pre-trained with public Calgary brain images and then fine-tuned for use in (1) patients with tumors in our center; (2) different anatomies, including knee and liver; (3) different k-space sampling masks with acceleration factors (AFs) of 2 and 6. As for the brain tumor dataset, the transfer learning results could remove the artifacts found in PI-GAN and yield smoother brain edges. The transfer learning results for the knee and liver were superior to those of the PI-GAN model trained with its own dataset using a smaller number of training cases. However, the learning procedure converged more slowly in the knee datasets compared to the learning in the brain tumor datasets. The reconstruction performance was improved by transfer learning both in the models with AFs of 2 and 6. Of these two models, the one with AF=2 showed better results. The results also showed that transfer learning with the pre-trained model could solve the problem of inconsistency between the training and test datasets and facilitate generalization to unseen data.
53.Cardiac Functional Analysis with Cine MRI via Deep Learning Reconstruction ⬇️
Retrospectively gated cine (retro-cine) MRI is the clinical standard for cardiac functional analysis. Deep learning (DL) based methods have been proposed for the reconstruction of highly undersampled MRI data and show superior image quality and magnitude faster reconstruction time than CS-based methods. Nevertheless, it remains unclear whether DL reconstruction is suitable for cardiac function analysis. To address this question, in this study we evaluate and compare the cardiac functional values (EDV, ESV and EF for LV and RV, respectively) obtained from highly accelerated MRI acquisition using DL based reconstruction algorithm (DL-cine) with values from CS-cine and conventional retro-cine. To the best of our knowledge, this is the first work to evaluate the cine MRI with deep learning reconstruction for cardiac function analysis and compare it with other conventional methods. The cardiac functional values obtained from cine MRI with deep learning reconstruction are consistent with values from clinical standard retro-cine MRI.
54.COVID-19 Lung Lesion Segmentation Using a Sparsely Supervised Mask R-CNN on Chest X-rays Automatically Computed from Volumetric CTs ⬇️
Chest X-rays of coronavirus disease 2019 (COVID-19) patients are frequently obtained to determine the extent of lung disease and are a valuable source of data for creating artificial intelligence models. Most work to date assessing disease severity on chest imaging has focused on segmenting computed tomography (CT) images; however, given that CTs are performed much less frequently than chest X-rays for COVID-19 patients, automated lung lesion segmentation on chest X-rays could be clinically valuable. There currently exists a universal shortage of chest X-rays with ground truth COVID-19 lung lesion annotations, and manually contouring lung opacities is a tedious, labor-intensive task. To accelerate severity detection and augment the amount of publicly available chest X-ray training data for supervised deep learning (DL) models, we leverage existing annotated CT images to generate frontal projection "chest X-ray" images for training COVID-19 chest X-ray models. In this paper, we propose an automated pipeline for segmentation of COVID-19 lung lesions on chest X-rays comprised of a Mask R-CNN trained on a mixed dataset of open-source chest X-rays and coronal X-ray projections computed from annotated volumetric CTs. On a test set containing 40 chest X-rays of COVID-19 positive patients, our model achieved IoU scores of 0.81
$\pm$ 0.03 and 0.79$\pm$ 0.03 when trained on a dataset of 60 chest X-rays and on a mixed dataset of 10 chest X-rays and 50 projections from CTs, respectively. Our model far outperforms current baselines with limited supervised training and may assist in automated COVID-19 severity quantification on chest X-rays.
55.Unsupervised MRI Reconstruction via Zero-Shot Learned Adversarial Transformers ⬇️
Supervised deep learning has swiftly become a workhorse for accelerated MRI in recent years, offering state-of-the-art performance in image reconstruction from undersampled acquisitions. Training deep supervised models requires large datasets of undersampled and fully-sampled acquisitions typically from a matching set of subjects. Given scarce access to large medical datasets, this limitation has sparked interest in unsupervised methods that reduce reliance on fully-sampled ground-truth data. A common framework is based on the deep image prior, where network-driven regularization is enforced directly during inference on undersampled acquisitions. Yet, canonical convolutional architectures are suboptimal in capturing long-range relationships, and randomly initialized networks may hamper convergence. To address these limitations, here we introduce a novel unsupervised MRI reconstruction method based on zero-Shot Learned Adversarial TransformERs (SLATER). SLATER embodies a deep adversarial network with cross-attention transformer blocks to map noise and latent variables onto MR images. This unconditional network learns a high-quality MRI prior in a self-supervised encoding task. A zero-shot reconstruction is performed on undersampled test data, where inference is performed by optimizing network parameters, latent and noise variables to ensure maximal consistency to multi-coil MRI data. Comprehensive experiments on brain MRI datasets clearly demonstrate the superior performance of SLATER against several state-of-the-art unsupervised methods.
56.A parameter refinement method for Ptychography based on Deep Learning concepts ⬇️
X-ray Ptychography is an advanced computational microscopy technique which is delivering exceptionally detailed quantitative imaging of biological and nanotechnology specimens. However coarse parametrisation in propagation distance, position errors and partial coherence frequently menaces the experiment viability. In this work we formally introduced these actors, solving the whole reconstruction as an optimisation problem. A modern Deep Learning framework is used to correct autonomously the setup incoherences, thus improving the quality of a ptychography reconstruction. Automatic procedures are indeed crucial to reduce the time for a reliable analysis, which has a significant impact on all the fields that use this kind of microscopy. We implemented our algorithm in our software framework, SciComPty, releasing it as open-source. We tested our system on both synthetic datasets and also on real data acquired at the TwinMic beamline of the Elettra synchrotron facility.