Search

In this paper, we propose a self-supervised single-view pixel-level accurate depth estimation network, called PLADE-Net. The PLADE-Net is the first work that shows unprecedented accuracy levels, exceeding 95\% in terms of the $\delta^1$ metric on the challenging KITTI dataset. Our PLADE-Net is based on a new network architecture with neural positional encoding and a novel loss function that borrows from the closed-form solution of the matting Laplacian to learn pixel-level accurate depth estimation from stereo images. Neural positional encoding allows our PLADE-Net to obtain more consistent depth estimates by letting the network reason about location-specific image properties such as lens and projection distortions. Our novel distilled matting Laplacian loss allows our network to predict sharp depths at object boundaries and more consistent depths in highly homogeneous regions. Our proposed method outperforms all previous self-supervised single-view depth estimation methods by a large margin on the challenging KITTI dataset, with unprecedented levels of accuracy. Furthermore, our PLADE-Net, naively extended for stereo inputs, outperforms the most recent self-supervised stereo methods, even without any advanced blocks like 1D correlations, 3D convolutions, or spatial pyramid pooling. We present extensive ablation studies and experiments that support our method's effectiveness on the KITTI, CityScapes, and Make3D datasets., Comment: Accepted paper (poster) at CVPR2021

With increasing fields of application for neural networks and the development of neural networks, the ability to explain deep learning models is also becoming increasingly important. Especially, prior to practical applications, it is crucial to analyze a model’s inference and the process of generating the results. A common explanation method is Class Activation Mapping(CAM) based method where it is often used to understand the last layer of the convolutional neural networks popular in the field of Computer Vision. In this paper, we propose a novel CAM method named Relevance-weighted Class Activation Mapping(Relevance-CAM) that utilizes Layer-wise Relevance Propagation to obtain the weighting components. This allows the explanation map to be faithful and robust to the shattered gradient problem, a shared problem of the gradient based CAM methods that causes noisy saliency maps for intermediate layers. Therefore, our proposed method can better explain a model by correctly analyzing the intermediate layers as well as the last convolutional layer. In this paper, we visualize how each layer of the popular image processing models extracts class specific features using Relevance-CAM, evaluate the localization ability, and show why the gradient based CAM cannot be used to explain the intermediate layers, proven by experimenting the weighting component. Relevance-CAM outperforms other CAM-based methods in recognition and localization evaluation in layers of any depth. The source code is available at: https://github.com/mongeoroo/Relevance-CAM

Channel pruning and tensor decomposition have received extensive attention in convolutional neural network compression. However, these two techniques are traditionally deployed in an isolated manner, leading to significant accuracy drop when pursuing high compression rates. In this paper, we propose a Collaborative Compression (CC) scheme, which joints channel pruning and tensor decomposition to compress CNN models by simultaneously learning the model sparsity and low-rankness. Specifically, we first investigate the compression sensitivity of each layer in the network, and then propose a Global Compression Rate Optimization that transforms the decision problem of compression rate into an optimization problem. After that, we propose multi-step heuristic compression to remove redundant compression units step-by-step, which fully considers the effect of the remaining compression space (i.e., unremoved compression units). Our method demonstrates superior performance gains over previous ones on various datasets and backbone architectures. For example, we achieve 52.9% FLOPs reduction by removing 48.4% parameters on ResNet-50 with only a Top-1 accuracy drop of 0.56% on ImageNet 2012., Comment: This paper is published in CVPR 2021

Pedestrian detection is used in many vision based applications ranging from video surveillance to autonomous driving. Despite achieving high performance, it is still largely unknown how well existing detectors generalize to unseen data. This is important because a practical detector should be ready to use in various scenarios in applications. To this end, we conduct a comprehensive study in this paper, using a general principle of direct cross-dataset evaluation. Through this study, we find that existing state-of-the-art pedestrian detectors, though perform quite well when trained and tested on the same dataset, generalize poorly in cross dataset evaluation. We demonstrate that there are two reasons for this trend. Firstly, their designs (e.g. anchor settings) may be biased towards popular benchmarks in the traditional single-dataset training and test pipeline, but as a result largely limit their generalization capability. Secondly, the training source is generally not dense in pedestrians and diverse in scenarios. Under direct cross-dataset evaluation, surprisingly, we find that a general purpose object detector, without pedestrian-tailored adaptation in design, generalizes much better compared to existing state-of-the-art pedestrian detectors. Furthermore, we illustrate that diverse and dense datasets, collected by crawling the web, serve to be an efficient source of pre-training for pedestrian detection. Accordingly, we propose a progressive training pipeline and find that it works well for autonomous-driving oriented pedestrian detection. Consequently, the study conducted in this paper suggests that more emphasis should be put on cross-dataset evaluation for the future design of generalizable pedestrian detectors. Code and models can be accessed at https://github.com/hasanirtiza/Pedestron., Comment: 10 pages, 1 figure

Convolutional neural network (CNN) has achieved unprecedented success in image super-resolution tasks in re-cent years. However, the network’s performance depends on the distribution of the training sets and degrades on out-of-distribution samples. This paper adopts a Bayesian approach for estimating uncertainty associated with output and applies it in a deep image super-resolution model to address the concern mentioned above. We use the uncertainty estimation technique using the batch-normalization layer, where stochasticity of the batch mean and variance generate Monte-Carlo (MC) samples. The MC samples, which are nothing but different super-resolved images using different stochastic parameters, reconstruct the image, and provide a confidence or uncertainty map of the reconstruction. We propose a faster approach for MC sample generation, and it allows the variable image size during testing. Therefore, it will be useful for image reconstruction domain. Our experimental findings show that this uncertainty map strongly relates to the quality of reconstruction generated by the deep CNN model and explains its limitation. Furthermore, this paper proposes an approach to reduce the model’s uncertainty for an input image, and it helps to defend the adversarial attacks on the image super-resolution model. The proposed uncertainty reduction technique also improves the performance of the model for out-of-distribution test images. To the best of our knowledge, we are the first to propose an adversarial defense mechanism in any image reconstruction domain.

One of the key steps in Neural Architecture Search (NAS) is to estimate the performance of candidate architectures. Existing methods either directly use the validation performance or learn a predictor to estimate the performance. However, these methods can be either computationally expensive or very inaccurate, which may severely affect the search efficiency and performance. Moreover, as it is very difficult to annotate architectures with accurate performance on specific tasks, learning a promising performance predictor is often non-trivial due to the lack of labeled data. In this paper, we argue that it may not be necessary to estimate the absolute performance for NAS. On the contrary, we may need only to understand whether an architecture is better than a baseline one. However, how to exploit this comparison information as the reward and how to well use the limited labeled data remains two great challenges. In this paper, we propose a novel Contrastive Neural Architecture Search (CTNAS) method which performs architecture search by taking the comparison results between architectures as the reward. Specifically, we design and learn a Neural Architecture Comparator (NAC) to compute the probability of candidate architectures being better than a baseline one. Moreover, we present a baseline updating scheme to improve the baseline iteratively in a curriculum learning manner. More critically, we theoretically show that learning NAC is equivalent to optimizing the ranking over architectures. Extensive experiments in three search spaces demonstrate the superiority of our CTNAS over existing methods., Accpeted by CVPR 2021. The code is available at https://github.com/chenyaofo/CTNAS

Consensus maximisation (MaxCon), which is widely used for robust fitting in computer vision, aims to find the largest subset of data that fits the model within some tolerance level. In this paper, we outline the connection between MaxCon problem and the abstract problem of finding the maximum upper zero of a Monotone Boolean Function (MBF) defined over the Boolean Cube. Then, we link the concept of influences (in a MBF) to the concept of outlier (in MaxCon) and show that influences of points belonging to the largest structure in data would generally be smaller under certain conditions. Based on this observation, we present an iterative algorithm to perform consensus maximisation. Results for both synthetic and real visual data experiments show that the MBF based algorithm is capable of generating a near optimal solution relatively quickly. This is particularly important where there are large number of outliers (gross or pseudo) in the observed data., To appear in CVPR 2021 as an ORAL paper. arXiv admin note: text overlap with arXiv:2005.05490

Few-shot object detection has made substantial progressby representing novel class objects using the feature representation learned upon a set of base class objects. However,an implicit contradiction between novel class classification and representation is unfortunately ignored. On the one hand, to achieve accurate novel class classification, the distributions of either two base classes must be far away fromeach other (max-margin). On the other hand, to precisely represent novel classes, the distributions of base classes should be close to each other to reduce the intra-class distance of novel classes (min-margin). In this paper, we propose a class margin equilibrium (CME) approach, with the aim to optimize both feature space partition and novel class reconstruction in a systematic way. CME first converts the few-shot detection problem to the few-shot classification problem by using a fully connected layer to decouple localization features. CME then reserves adequate margin space for novel classes by introducing simple-yet-effective class margin loss during feature learning. Finally, CME pursues margin equilibrium by disturbing the features of novel class instances in an adversarial min-max fashion. Experiments on Pascal VOC and MS-COCO datasets show that CME significantly improves upon two baseline detectors (up to $3\sim 5\%$ in average), achieving state-of-the-art performance. Code is available at https://github.com/Bohao-Lee/CME ., Comment: This paper has been modified by the author due to errors

We wish to detect specific categories of objects, for online vision systems that will run in the real world. Object detection is already very challenging. It is even harder when the images are blurred, from the camera being in a car or a hand-held phone. Most existing efforts either focused on sharp images, with easy to label ground truth, or they have treated motion blur as one of many generic corruptions. Instead, we focus especially on the details of egomotion induced blur. We explore five classes of remedies, where each targets different potential causes for the performance gap between sharp and blurred images. For example, first deblurring an image changes its human interpretability, but at present, only partly improves object detection. The other four classes of remedies address multi-scale texture, out-of-distribution testing, label generation, and conditioning by blur-type. Surprisingly, we discover that custom label generation aimed at resolving spatial ambiguity, ahead of all others, markedly improves object detection. Also, in contrast to findings from classification, we see a noteworthy boost by conditioning our model on bespoke categories of motion blur. We validate and cross-breed the different remedies experimentally on blurred COCO images and real-world blur datasets, producing an easy and practical favorite model with superior detection rates., Comment: Mirroring accepted CVPR paper. Added results for other real-world blur datasets. Main paper: 8 pages + 3 references. Supplemental: 9 pages

OCR-based image captioning aims to automatically describe images based on all the visual entities (both visual objects and scene text) in images. Compared with conventional image captioning, the reasoning of scene text is required for OCR-based image captioning since the generated descriptions often contain multiple OCR tokens. Existing methods attempt to achieve this goal via encoding the OCR tokens with rich visual and semantic representations. However, strong correlations between OCR tokens may not be established with such limited representations. In this paper, we propose to enhance the connections between OCR tokens from the viewpoint of exploiting the geometrical relationship. We comprehensively consider the height, width, distance, IoU and orientation relations between the OCR tokens for constructing the geometrical relationship. To integrate the learned relation as well as the visual and semantic representations into a unified framework, a Long Short-Term Memory plus Relation-aware pointer network (LSTM-R) architecture is presented in this paper. Under the guidance of the geometrical relationship between OCR tokens, our LSTM-R capitalizes on a newly-devised relation-aware pointer network to select OCR tokens from the scene text for OCR-based image captioning. Extensive experiments demonstrate the effectiveness of our LSTM-R. More remarkably, LSTM-R achieves state-of-the-art performance on TextCaps, with the CIDEr-D score being increased from 98.0% to 109.3%.

In this paper, we diagnose deep neural networks for 3D point cloud processing to explore utilities of different intermediate-layer network architectures. We propose a number of hypotheses on the effects of specific intermediate-layer network architectures on the representation capacity of DNNs. In order to prove the hypotheses, we design five metrics to diagnose various types of DNNs from the following perspectives, information discarding, information concentration, rotation robustness, adversarial robustness, and neighborhood inconsistency. We conduct comparative studies based on such metrics to verify the hypotheses. We further use the verified hypotheses to revise intermediate-layer architectures of existing DNNs and improve their utilities. Experiments demonstrate the effectiveness of our method. The code will be released when this paper is accepted.

Deep models have shown their vulnerability when processing adversarial samples. As for the black-box attack, without access to the architecture and weights of the attacked model, training a substitute model for adversarial attacks has attracted wide attention. Previous substitute training approaches focus on stealing the knowledge of the target model based on real training data or synthetic data, without exploring what kind of data can further improve the transferability between the substitute and target models. In this paper, we propose a novel perspective substitute training that focuses on designing the distribution of data used in the knowledge stealing process. More specifically, a diverse data generation module is proposed to synthesize large-scale data with wide distribution. And adversarial substitute training strategy is introduced to focus on the data distributed near the decision boundary. The combination of these two modules can further boost the consistency of the substitute model and target model, which greatly improves the effectiveness of adversarial attack. Extensive experiments demonstrate the efficacy of our method against state-of-the-art competitors under non-target and target attack settings. Detailed visualization and analysis are also provided to help understand the advantage of our method., Comment: 10 pages, 6 figures, 6 tables, 1 algorithm, To appear in CVPR 2021 as a poster paper

This paper tackles the task of Few-Shot Video Object Segmentation (FSVOS), i.e., segmenting objects in the query videos with certain class specified in a few labeled support images. The key is to model the relationship between the query videos and the support images for propagating the object information. This is a many-to-many problem and often relies on full-rank attention, which is computationally intensive. In this paper, we propose a novel Domain Agent Network (DAN), breaking down the full-rank attention into two smaller ones. We consider one single frame of the query video as the domain agent, bridging between the support images and the query video. Our DAN allows a linear space and time complexity as opposed to the original quadratic form with no loss of performance. In addition, we introduce a learning strategy by combining meta-learning with online learning to further improve the segmentation accuracy. We build a FSVOS benchmark on the Youtube-VIS dataset and conduct experiments to demonstrate that our method outperforms baselines on both computational cost and accuracy, achieving the state-of-the-art performance. Code is available at https://github.com/scutpaul/DANet.

Unsupervised domain adaptation (UDA) aims to improve the classification performance on an unlabeled target domain by leveraging information from a fully labeled source domain. Recent approaches explore domain-invariant and class-discriminant representations to tackle this task. These methods, however, ignore the interaction between domain alignment learning and class discrimination learning. As a result, the missing or inadequate tradeoff between domain alignment and class discrimination are prone to the problem of negative transfer. In this paper, we propose Dynamic Weighted Learning (DWL) to avoid the discriminability vanishing problem caused by excessive alignment learning and domain misalignment problem caused by excessive discriminant learning. Technically, DWL dynamically weights the learning losses of alignment and discriminability by introducing the degree of alignment and discriminability. Besides, the problem of sample imbalance across domains is first considered in our work, and we solve the problem by weighing the samples to guarantee information balance across domains. Extensive experiments demonstrate that DWL has an excellent performance in several benchmark datasets., Comment: This paper has been accepted by CVPR2021

Graph neural networks (GNNs) emerged recently as a standard toolkit for learning from data on graphs. Current GNN designing works depend on immense human expertise to explore different message-passing mechanisms, and require manual enumeration to determine the proper message-passing depth. Inspired by the strong searching capability of neural architecture search (NAS) in CNN, this paper proposes Graph Neural Architecture Search (GNAS) with novel-designed search space. The GNAS can automatically learn better architecture with the optimal depth of message passing on the graph. Specifically, we design Graph Neural Architecture Paradigm (GAP) with tree-topology computation procedure and two types of fine-grained atomic operations (feature filtering and neighbor aggregation) from message-passing mechanism to construct powerful graph network search space. Feature filtering performs adaptive feature selection, and neighbor aggregation captures structural information and calculates neighbors' statistics. Experiments show that our GNAS can search for better GNNs with multiple message-passing mechanisms and optimal message-passing depth. The searched network achieves remarkable improvement over state-of-the-art manual designed and search-based GNNs on five large-scale datasets at three classical graph tasks. Codes can be found at https://github.com/phython96/GNAS-MP., This paper has been accepted by CVPR2021

In this paper, we tackle the problem of convolutional neural network design. Instead of focusing on the design of the overall architecture, we investigate a design space that is usually overlooked, i.e. adjusting the channel configurations of predefined networks. We find that this adjustment can be achieved by shrinking widened baseline networks and leads to superior performance. Based on that, we articulate the heterogeneity hypothesis: with the same training protocol, there exists a layer-wise differentiated network architecture (LW-DNA) that can outperform the original network with regular channel configurations but with a lower level of model complexity. The LW-DNA models are identified without extra computational cost or training time compared with the original network. This constraint leads to controlled experiments which direct the focus to the importance of layer-wise specific channel configurations. LW-DNA models come with advantages related to overfitting, i.e. the relative relationship between model complexity and dataset size. Experiments are conducted on various networks and datasets for image classification, visual tracking and image restoration. The resultant LW-DNA models consistently outperform the baseline models. Code is available at https://github.com/ofsoundof/Heterogeneity_Hypothesis., CVPR2021 paper

Super-resolution based on deep convolutional networks is currently gaining much attention from both academia and industry. However, lack of proper evaluation measures makes it difficult to compare approaches, hampering progress in the field. Traditional measures, such as PSNR or SSIM, are known to poorly correlate with the human perception of image quality. Therefore, in existing works common practice is also to report Mean-Opinion-Score (MOS) — the results of human evaluation of super-resolved images. Unfortunately, the MOS values from different papers are not directly comparable, due to the varying number of raters, their subjectivity, etc. By this paper, we introduce Neural Side-By-Side — a new measure that allows super-resolution models to be compared automatically, effectively approximating human preferences. Namely, we collect a large dataset of aligned image pairs, which were produced by different super-resolution models. Then each pair is annotated by several raters, who were instructed to choose a more visually appealing image. Given the dataset and the labels, we trained a CNN model that obtains a pair of images and for each image predicts a probability of being more preferable than its counterpart. In this work, we show that Neural Side-By-Side generalizes across both new models and new data. Hence, it can serve as a natural approximation of human preferences, which can be used to compare models or tune hyperparameters without raters’ assistance. We open-source the dataset and the pretrained model1 and expect that it will become a handy tool for researchers and practitioners.

Convolutional neural networks have made remarkable progress in the face recognition field. The more the technology of face recognition advances, the greater discriminative features into a face template. However, this increases the threat to user privacy in case the template is exposed. In this paper, we present a modular architecture for face template protection, called IronMask, that can be combined with any face recognition system using angular distance metric. We circumvent the need for binarization, which is the main cause of performance degradation in most existing face template protections, by proposing a new real-valued error-correcting-code that is compatible with real-valued templates and can therefore, minimize performance degradation. We evaluate the efficacy of IronMask by extensive experiments on two face recognitions, ArcFace and CosFace with three datasets, CMU-Multi-PIE, FEI, and Color-FERET. According to our experimental results, IronMask achieves a true accept rate (TAR) of 99.79% at a false accept rate (FAR) of 0.0005% when combined with ArcFace, and 95.78% TAR at 0% FAR with CosFace, while providing at least 115-bit security against known attacks., The submission is a 13 pages of paper which consists of 3 figures, 3 tables. It is the full version of CVPR '21 paper (The Conference on Computer Vision and Patter Recognition)

Unsupervised Domain Adaptation (UDA) aims to generalize the knowledge learned from a well-labeled source domain to an unlabeled target domain. Recently, adversarial domain adaptation with two distinct classifiers (bi-classifier) has been introduced into UDA which is effective to align distributions between different domains. Previous bi-classifier adversarial learning methods only focus on the similarity between the outputs of two distinct classifiers. However, the similarity of the outputs cannot guarantee the accuracy of target samples, i.e., target samples may match to wrong categories even if the discrepancy between two classifiers is small. To challenge this issue, in this paper, we propose a cross-domain gradient discrepancy minimization (CGDM) method which explicitly minimizes the discrepancy of gradients generated by source samples and target samples. Specifically, the gradient gives a cue for the semantic information of target samples so it can be used as a good supervision to improve the accuracy of target samples. In order to compute the gradient signal of target samples, we further obtain target pseudo labels through a clustering-based self-supervised learning. Extensive experiments on three widely used UDA datasets show that our method surpasses many previous state-of-the-arts. Codes are available at https://github.com/lijin118/CGDM., Accepted to CVPR 2021, Codes are avaliable at https://github.com/lijin118/CGDM

Quantifying the importance of each training point to a learning task is a fundamental problem in machine learning and the estimated importance scores have been leveraged to guide a range of data workflows such as data summarization and domain adaption. One simple idea is to use the leave-one-out error of each training point to indicate its importance. Recent work has also proposed to use the Shapley value, as it defines a unique value distribution scheme that satisfies a set of appealing properties. However, calculating Shapley values is often expensive, which limits its applicability in real-world applications at scale. Multiple heuristics to improve the scalability of calculating Shapley values have been proposed recently, with the potential risk of compromising their utility in real-world applications. \textit{How well do existing data quantification methods perform on existing workflows? How do these methods compare with each other, empirically and theoretically? Must we sacrifice scalability for the utility in these workflows when using these methods?} In this paper, we conduct a novel theoretical analysis comparing the utility of different importance quantification methods, and report extensive experimental studies on existing and proposed workflows such as noisy label detection, watermark removal, data summarization, data acquisition, and domain adaptation. We show that Shapley value approximation based on a $K$NN surrogate over pre-trained feature embeddings obtains comparable utility with existing algorithms while achieving significant scalability improvement, often by orders of magnitude. Our theoretical analysis also justifies its advantage over the leave-one-out error. The code is available at \url{https://github.com/AI-secure/Shapley-Study}.

Although existing fully-supervised defocus blur detection (DBD) models significantly improve performance, training such deep models requires abundant pixel-level manual annotation, which is highly time-consuming and error-prone. Addressing this issue, this paper makes an effort to train a deep DBD model without using any pixel-level annotation. The core insight is that a defocus blur region/focused clear area can be arbitrarily pasted to a given realistic full blurred image/full clear image without affecting the judgment of the full blurred image/full clear image. Specifically, we train a generator G in an adversarial manner against dual discriminators D c and D b . G learns to produce a DBD mask that generates a composite clear image and a composite blurred image through copying the focused area and unfocused region from corresponding source image to another full clear image and full blurred image. Then, D c and D b can not distinguish them from realistic full clear image and full blurred image simultaneously, achieving a self-generated DBD by an implicit manner to define what a defocus blur area is. Besides, we propose a bilateral triplet-excavating constraint to avoid the degenerate problem caused by the case one discriminator defeats the other one. Comprehensive experiments on two widely-used DBD datasets demonstrate the superiority of the proposed approach. Source codes are available at: https://github.com/shangcai1/SG.

Recent works on localization and mapping from privacy preserving line features have made significant progress towards addressing the privacy concerns arising from cloud-based solutions in mixed reality and robotics. The requirement for calibrated cameras is a fundamental limitation for these approaches, which prevents their application in many crowd-sourced mapping scenarios. In this paper, we propose a solution to the uncalibrated privacy preserving localization and mapping problem. Our approach simultaneously recovers the intrinsic and extrinsic calibration of a camera from line-features only. This enables uncalibrated devices to both localize themselves within an existing map as well as contribute to the map, while preserving the privacy of the image contents. Furthermore, we also derive a solution to bootstrapping maps from scratch using only uncalibrated devices. Our approach provides comparable performance to the calibrated scenario and the privacy compromising alternatives based on traditional point features.

Hand gesture-to-gesture translation is a significant and interesting problem, which serves as a key role in many applications, such as sign language production. This task involves fine-grained structure understanding of the mapping between the source and target gestures. Current works follow a data-driven paradigm based on sparse 2D joint representation. However, given the insufficient representation capability of 2D joints, this paradigm easily leads to blurry generation results with incorrect structure. In this paper, we propose a novel model-aware gesture-to-gesture translation framework, which introduces hand prior with hand meshes as the intermediate representation. To take full advantage of the structured hand model, we first build a dense topology map aligning the image plane with the encoded embedding of the visible hand mesh. Then, a transformation flow is calculated based on the correspondence of the source and target topology map. During the generation stage, we inject the topology information into generation streams by modulating the activations in a spatially-adaptive manner. Further, we incorporate the source local characteristic to enhance the translated gesture image according to the transformation flow. Extensive experiments on two benchmark datasets have demonstrated that our method achieves new state-of-the-art performance.

In this paper, we introduce Coarse-Fine Networks, a two-stream architecture which benefits from different abstractions of temporal resolution to learn better video representations for long-term motion. Traditional Video models process inputs at one (or few) fixed temporal resolution without any dynamic frame selection. However, we argue that, processing multiple temporal resolutions of the input and doing so dynamically by learning to estimate the importance of each frame can largely improve video representations, specially in the domain of temporal activity localization. To this end, we propose (1) Grid Pool, a learned temporal downsampling layer to extract coarse features, and, (2) Multi-stage Fusion, a spatio-temporal attention mechanism to fuse a fine-grained context with the coarse features. We show that our method outperforms the state-of-the-arts for action detection in public datasets including Charades with a significantly reduced compute and memory footprint. The code is available at https://github.com/kkahatapitiya/Coarse-Fine-Networks, Comment: To appear at CVPR 2021

We investigate the problem of generating 3D meshes from single free-hand sketches, aiming at fast 3D modeling for novice users. It can be regarded as a single-view reconstruction problem, but with unique challenges, brought by the variation and conciseness of sketches. Ambiguities in poorly-drawn sketches could make it hard to determine how the sketched object is posed. In this paper, we address the importance of viewpoint specification for overcoming such ambiguities, and propose a novel view-aware generation approach. By explicitly conditioning the generation process on a given viewpoint, our method can generate plausible shapes automatically with predicted viewpoints, or with specified viewpoints to help users better express their intentions. Extensive evaluations on various datasets demonstrate the effectiveness of our view-aware design in solving sketch ambiguities and improving reconstruction quality., Comment: Accepted to CVPR 2021

An effective and efficient architecture performance evaluation scheme is essential for the success of Neural Architecture Search (NAS). To save computational cost, most of existing NAS algorithms often train and evaluate intermediate neural architectures on a small proxy dataset with limited training epochs. But it is difficult to expect an accurate performance estimation of an architecture in such a coarse evaluation way. This paper advocates a new neural architecture evaluation scheme, which aims to determine which architecture would perform better instead of accurately predict the absolute architecture performance. Therefore, we propose a relativistic architecture performance predictor in NAS (ReNAS). We encode neural architectures into feature tensors, and further refining the representations with the predictor. The proposed relativistic performance predictor can be deployed in discrete searching methods to search for the desired architectures without additional evaluation. Experimental results on NAS-Bench-101 dataset suggests that, sampling 424 (0.1% of the entire search space) neural architectures and their corresponding validation performance is already enough for learning an accurate architecture performance predictor. The accuracies of our searched neural architectures on NAS-Bench-101 and NAS-Bench-201 datasets are higher than that of the state-of-the-art methods and show the priority of the proposed method.

This paper introduces a method to encode the blur operators of an arbitrary dataset of sharp-blur image pairs into a blur kernel space. Assuming the encoded kernel space is close enough to in-the-wild blur operators, we propose an alternating optimization algorithm for blind image deblurring. It approximates an unseen blur operator by a kernel in the encoded space and searches for the corresponding sharp image. Unlike recent deep-learning-based methods, our system can handle unseen blur kernel, while avoiding using complicated handcrafted priors on the blur operator often found in classical methods. Due to the method’s design, the encoded kernel space is fully differentiable, thus can be easily adopted in deep neural network models. Moreover, our method can be used for blur synthesis by transferring existing blur operators from a given dataset into a new domain. Finally, we provide experimental results to confirm the effectiveness of the proposed method. The code is available at https://github.com/VinAIResearch/blur-kernelspace-exploring.

Unsupervised domain adaptation (UDA) enables transferring knowledge from a related source domain to a fully unlabeled target domain. Despite the significant advances in UDA, the performance gap remains quite large between UDA and supervised learning with fully labeled target data. Active domain adaptation (ADA) mitigates the gap under minimal annotation cost by selecting a small quota of target samples to annotate and incorporating them into training. Due to the domain shift, the query selection criteria of prior active learning methods may be ineffective to select the most informative target samples for annotation. In this paper, we propose Transferable Query Selection (TQS), which selects the most informative samples under domain shift by an ensemble of three new criteria: transferable committee, transferable uncertainty, and transferable domainness. We further develop a randomized selection algorithm to enhance the diversity of the selected samples. Experiments show that TQS remarkably outperforms previous UDA and ADA methods on several domain adaptation datasets. Deeper analyses demonstrate that TQS can select the most informative target samples under the domain shift.

Panoptic segmentation aims to partition an image to object instances and semantic content for thing and stuff categories, respectively. To date, learning weakly supervised panoptic segmentation (WSPS) with only image-level labels remains unexplored. In this paper, we propose an efficient jointly thing-and-stuff mining (JTSM) framework for WSPS. To this end, we design a novel mask of interest pooling (MoIPool) to extract fixed-size pixel-accurate feature maps of arbitrary-shape segmentations. MoIPool enables a panoptic mining branch to leverage multiple instance learning (MIL) to recognize things and stuff segmentation in a unified manner. We further refine segmentation masks with parallel instance and semantic segmentation branches via self-training, which collaborates the mined masks from panoptic mining with bottom-up object evidence as pseudo-ground-truth labels to improve spatial coherence and contour localization. Experimental results demonstrate the effectiveness of JTSM on PASCAL VOC and MS COCO. As a by-product, we achieve competitive results for weakly supervised object detection and instance segmentation. This work is a first step towards tackling challenge panoptic segmentation task with only image-level labels.

Domain adaptation (DA) enables knowledge transfer from a labeled source domain to an unlabeled target domain by reducing the cross-domain distribution discrepancy. Most prior DA approaches leverage complicated and powerful deep neural networks to improve the adaptation capacity and have shown remarkable success. However, they may have a lack of applicability to real-world situations such as real-time interaction, where low target inference latency is an essential requirement under limited computational budget. In this paper, we tackle the problem by proposing a dynamic domain adaptation (DDA) framework, which can simultaneously achieve efficient target inference in low-resource scenarios and inherit the favorable cross-domain generalization brought by DA. In contrast to static models, as a simple yet generic method, DDA can integrate various domain confusion constraints into any typical adaptive network, where multiple intermediate classifiers can be equipped to infer "easier" and "harder" target data dynamically. Moreover, we present two novel strategies to further boost the adaptation performance of multiple prediction exits: 1) a confidence score learning strategy to derive accurate target pseudo labels by fully exploring the prediction consistency of different classifiers; 2) a class-balanced self-training strategy to explicitly adapt multi-stage classifiers from source to target without losing prediction diversity. Extensive experiments on multiple benchmarks are conducted to verify that DDA can consistently improve the adaptation performance and accelerate target inference under domain shift and limited resources scenarios, Accepted by CVPR 2021

Specular highlight detection and removal are fundamental and challenging tasks. Although recent methods have achieved promising results on the two tasks by training on synthetic training data in a supervised manner, they are typically solely designed for highlight detection or removal, and their performance usually deteriorates significantly on real-world images. In this paper, we present a novel network that aims to detect and remove highlights from natural images. To remove the domain gap between synthetic training samples and real test images, and support the investigation of learning-based approaches, we first introduce a dataset with about 16K real images, each of which has the corresponding ground truths of highlight detection and removal. Using the presented dataset, we develop a multi-task network for joint highlight detection and removal, based on a new specular highlight image formation model. Experiments on the benchmark datasets and our new dataset show that our approach clearly outperforms state-of-the-art methods for both highlight detection and removal.

In this paper, we consider stochastic second-order methods for minimizing a finite summation of nonconvex functions. One important key is to find an ingenious but cheap scheme to incorporate local curvature information. Since the true Hessian matrix is often a combination of a cheap part and an expensive part, we propose a structured stochastic quasi-Newton method by using partial Hessian information as much as possible. By further exploiting either the low-rank structure or the kronecker-product properties of the quasi-Newton approximations, the computation of the quasi-Newton direction is affordable. Global convergence to stationary point and local superlinear convergence rate are established under some mild assumptions. Numerical results on logistic regression, deep autoencoder networks and deep convolutional neural networks show that our proposed method is quite competitive to the state-of-the-art methods.

Message passing neural network has been an effective method to represent dependencies among nodes by propagating messages. However, most of message passing algorithms focus on one structure and messages are estimated by one single approach. For real-world data, like facial action units (AUs), the dependencies may vary in terms of different expressions and individuals. In this paper, we propose a novel hybrid message passing neural network with performance-driven structures (HMP-PS), which combines complementary message passing methods and captures more possible structures in a Bayesian manner. Particularly, a performance-driven Monte Carlo Markov Chain sampling method is proposed for generating high performance graph structures. Besides, hybrid message passing is proposed to combine different types of messages, which provide the complementary information. The contribution of each type of message is adaptively adjusted along with different inputs. The experiments on two widely used benchmark datasets, i.e., BP4D and DISFA, validate that our proposed method can achieve the state-of-the-art performance.

Deep learning based object detection approaches have achieved great progress with the benefit from large amount of labeled images. However, image annotation remains a laborious, time-consuming and error-prone process. To further improve the performance of detectors, we seek to exploit all available labeled data and excavate useful samples from massive unlabeled images in the wild, which is rarely discussed before. In this paper, we present a positive-unlabeled learning based scheme to expand training data by purifying valuable images from massive unlabeled ones, where the original training data are viewed as positive data and the unlabeled images in the wild are unlabeled data. To effectively utilized these purified data, we propose a self-distillation algorithm based on hint learning and ground truth bounded knowledge distillation. Experimental results verify that the proposed positive-unlabeled data purification can strengthen the original detector by mining the massive unlabeled data. In particular, our method boosts the mAP of FPN by +2.0% on COCO benchmark.

In this paper, we aim to synthesize cell microscopy images under different molecular interventions, motivated by practical applications to drug development. Building on the recent success of graph neural networks for learning molecular embeddings and flow-based models for image generation, we propose Mol2Image: a flow-based generative model for molecule to cell image synthesis. To generate cell features at different resolutions and scale to high-resolution images, we develop a novel multi-scale flow architecture based on a Haar wavelet image pyramid. To maximize the mutual information between the generated images and the molecular interventions, we devise a training strategy based on contrastive learning. To evaluate our model, we propose a new set of metrics for biological image generation that are robust, interpretable, and relevant to practitioners. We show quantitatively that our method learns a meaningful embedding of the molecular intervention, which is translated into an image representation reflecting the biological effects of the intervention.

Traditional methods for Unsupervised Domain Adaptation (UDA) targeting semantic segmentation exploit information common to the source and target domains, using both labeled source data and unlabeled target data. In this paper, we investigate a setting where the source data is un-available, but the classifier trained on the source data is; hence named "model adaptation". Such a scenario arises when data sharing is prohibited, for instance, because of privacy, or Intellectual Property (IP) issues.To tackle this problem, we propose a method that reduces the uncertainty of predictions on the target domain data. We accomplish this in two ways: minimizing the entropy of the predicted posterior, and maximizing the noise robustness of the feature representation. We show the efficacy of our method on the transfer of segmentation from computer generated images to real-world driving images, and transfer between data collected in different cities, and surprisingly reach performance comparable with that of the methods that have access to source data.

Unsupervised domain adaptation (UDA) methods for learning domain invariant representations have achieved remarkable progress. However, most of the studies were based on direct adaptation from the source domain to the target domain and have suffered from large domain discrepancies. In this paper, we propose a UDA method that effectively handles such large domain discrepancies. We introduce a fixed ratio-based mixup to augment multiple intermediate domains between the source and target domain. From the augmented-domains, we train the source-dominant model and the target-dominant model that have complementary characteristics. Using our confidence-based learning methodologies, e.g., bidirectional matching with high-confidence predictions and self-penalization using low-confidence predictions, the models can learn from each other or from its own results. Through our proposed methods, the models gradually transfer domain knowledge from the source to the target domain. Extensive experiments demonstrate the superiority of our proposed method on three public benchmarks: Office-31, Office-Home, and VisDA-2017., Accepted to CVPR 2021

Image and video descriptors are an omnipresent tool in computer vision and its application fields like mobile robotics. Many hand-crafted and in particular learned image descriptors are numerical vectors with a potentially (very) large number of dimensions. Practical considerations like memory consumption or time for comparisons call for the creation of compact representations. In this paper, we use hyperdimensional computing (HDC) as an approach to systematically combine information from a set of vectors in a single vector of the same dimensionality. HDC is a known technique to perform symbolic processing with distributed representations in numerical vectors with thousands of dimensions. We present a HDC implementation that is suitable for processing the output of existing and future (deep learning based) image descriptors. We discuss how this can be used as a framework to process descriptors together with additional knowledge by simple and fast vector operations. A concrete outcome is a novel HDC-based approach to aggregate a set of local image descriptors together with their image positions in a single holistic descriptor. The comparison to available holistic descriptors and aggregation methods on a series of standard mobile robotics place recognition experiments shows a 20% improvement in average performance and > 2× better worst-case performance compared to runner-up.

In this paper we study structure from motion problems for 1D radial cameras. Under this model the projection of a 3D point is a line in the image plane going through the principal point, which makes the model invariant to radial distortion and changes in focal length. It can therefore effectively be applied to uncalibrated image collections without the need for explicit estimation of camera intrinsics.We show that the reprojection errors of 1D radial cameras are examples of quasiconvex functions. This opens up the possibility to solve a general class of relevant reconstruction problems globally optimally using tools from convex optimization. In fact, our resulting algorithm is based on solving a series of LP problems. We perform an extensive experimental evaluation, on both synthetic and real data, showing that a whole class of multiview geometry problems across a range of different cameras models with varying and unknown intrinsic calibration can be reliably and accurately solved within the same framework.1

Recently, how to exploit unlabeled data for training face recognition models has been attracting increasing attention. However, few works consider the unlabeled shallow data1 in real-world scenarios. The existing semi-supervised face recognition methods that focus on generating pseudo labels or minimizing softmax classification probabilities of the unlabeled data do not work very well on the unlabeled shallow data. It is still a challenge on how to effectively utilize the unlabeled shallow face data to improve the performance of face recognition. In this paper, we propose a novel face recognition method, named VirFace, to effectively exploit the unlabeled shallow data for face recognition. VirFace consists of VirClass and VirInstance. Specifically, VirClass enlarges the inter-class distance by injecting the unlabeled data as new identities, while VirInstance produces virtual instances sampled from the learned distribution of each identity to further enlarge the inter-class distance. To the best of our knowledge, we are the first to tackle the problem of unlabeled shallow face data. Extensive experiments have been conducted on both the small- and large-scale datasets, e.g. LFW and IJB-C, etc, demonstrating the superiority of the proposed method.

In this paper, we aim to recognize materials with combined use of auditory and visual perception. To this end, we construct a new dataset named GLAudio that consists of both the geometry of the object being struck and the sound captured from either modal sound synthesis (for virtual objects) or real measurements (for real objects). Besides global geometries, our dataset also takes local geometries around different hitpoints into consideration. This local information is less explored in existing datasets. We demonstrate that local geometry has a greater impact on the sound than the global geometry and offers more cues in material recognition. To extract features from different modalities and perform proper fusion, we propose a new deep neural network GLAVNet that comprises multiple branches and a well-designed fusion module. Once trained on GLAudio, our GLAVNet provides state-of-the-art performance on material identification and supports fine-grained material categorization.

Acquisition and rendering of photo-realistic human heads is a highly challenging research problem of particular importance for virtual telepresence. Currently, the highest quality is achieved by volumetric approaches trained in a person-specific manner on multi-view data. These models better represent fine structure, such as hair, compared to simpler mesh-based models. Volumetric models typically employ a global code to represent facial expressions, such that they can be driven by a small set of animation parameters. While such architectures achieve impressive rendering quality, they can not easily be extended to the multi-identity setting. In this paper, we devise a novel approach for predicting volumetric avatars of the human head given just a small number of inputs. We enable generalization across identities by a novel parameterization that combines neural radiance fields with local, pixel-aligned features extracted directly from the inputs, thus side-stepping the need for very deep or complex networks. Our approach is trained in an end-to-end manner solely based on a photometric re-rendering loss without requiring explicit 3D supervision. We demonstrate that our approach outperforms the existing state of the art in terms of quality and is able to generate faithful facial expressions in a multi-identity setting.

These days, unsupervised super-resolution (SR) is soaring due to its practical and promising potential in real scenarios. The philosophy of off-the-shelf approaches lies in the augmentation of unpaired data, i.e. first generating synthetic low-resolution (LR) images ${\mathcal{Y}^g}$ corresponding to real-world high-resolution (HR) images ${\mathcal{X}^r}$ in the real-world LR domain ${\mathcal{Y}^r}$, and then utilizing the pseudo pairs $\left\{ {{\mathcal{Y}^g},{\mathcal{X}^r}} \right\}$ for training in a supervised manner. Unfortunately, since image translation itself is an extremely challenging task, the SR performance of these approaches is severely limited by the domain gap between generated synthetic LR images and real LR images. In this paper, we propose a novel domain-distance aware super-resolution (DASR) approach for unsupervised real-world image SR. The domain gap between training data (e.g. ${\mathcal{Y}^g}$) and testing data (e.g. ${\mathcal{Y}^r}$) is addressed with our domain-gap aware training and domain-distance weighted supervision strategies. Domain-gap aware training takes additional benefit from real data in the target domain while domain-distance weighted supervision brings forward the more rational use of labeled source domain data. The proposed method is validated on synthetic and real datasets and the experimental results show that DASR consistently outperforms state-of-the-art unsupervised SR approaches in generating SR outputs with more realistic and natural textures. Codes are available at https://github.com/ShuhangGu/DASR.

Previous online 3D dense reconstruction methods struggle to achieve the balance between memory storage and surface quality, largely due to the usage of stagnant underlying geometry representation, such as TSDF (truncated signed distance functions) or surfels, without any knowledge of the scene priors. In this paper, we present DI-Fusion (Deep Implicit Fusion), based on a novel 3D representation, i.e. Probabilistic Local Implicit Voxels (PLIVoxs), for online 3D reconstruction with a commodity RGB-D camera. Our PLIVox encodes scene priors considering both the local geometry and uncertainty parameterized by a deep neural network. With such deep priors, we are able to perform online implicit 3D reconstruction achieving state-of-the-art camera trajectory estimation accuracy and mapping quality, while achieving better storage efficiency compared with previous online 3D reconstruction approaches. Our implementation is available at https://www.github.com/huangjh-pub/di-fusion.

Visual Question Answering (VQA) models are notorious for their tendency to rely on dataset biases.The large and unbalanced diversity of questions and concepts involved in VQA and the lack of high standard annotated data tend to prevent models from learning to `reason', leading them to perform `educated guesses' instead, relying on specific training set statistics, which is not helpful for generalizing to real world scenarios.In this paper, we claim that the standard evaluation metric, which consists in measuring the overall in-domain accuracy is misleading. Questions and concepts being unequally distributed, it tends to favor models which exploit subtle training set statistics.Alternatively, naively evaluating generalization by introducing artificial distribution shift between train and test splits is also not completely satisfying. First, the shifts do not reflect real words tendencies, resulting in unsuitable models; second, since the shifts are artificially handcrafted, trained models are specifically designed for this particular setting, and paradoxically do not generalize to other configurations.We propose the GQA-OOD benchmark designed to overcome these concerns: we measure and compare accuracy over, both, rare and frequent question-answer pairs and argue that the former is better suited to the evaluation of reasoning abilities, which we experimentally validate with models trained to more or less exploit biases. In a large-scale study involving 7 VQA models and 3 bias reduction techniques, we also experimentally demonstrate that these models fail to address questions involving infrequent concepts and provide recommendations for future directions of research.

This paper presents an end-to-end neural network for multiple rotation averaging in SfM. Due to the manifold constraint of rotations, conventional methods usually take two separate steps involving spanning tree based initialization and iterative nonlinear optimization respectively. These methods can suffer from bad initializations due to the noisy spanning tree or outliers in input relative rotations. To handle these problems, we propose to integrate initialization and optimization together in an unified graph neural network via a novel differentiable multi-source propagation module. Specifically, our network utilizes the image context and geometric cues in feature correspondences to reduce the impact of outliers. Furthermore, unlike the methods that utilize the spanning tree to initialize orientations according to a single reference node in a top-down manner, our net-work initializes orientations according to multiple sources while utilizing information from all neighbors in a differentiable way. More importantly, our end-to-end formulation also enables iterative re-weighting of input relative orientations at test time to improve the accuracy of the final estimation by minimizing the impact of outliers. We demonstrate the effectiveness of our method on two real-world datasets, achieving state-of-the-art performance.

Temporal action localization is an important yet challenging task in video understanding. Typically, such a task aims at inferring both the action category and localization of the start and end frame for each action instance in a long, untrimmed video.While most current models achieve good results by using pre-defined anchors and numerous actionness, such methods could be bothered with both large number of outputs and heavy tuning of locations and sizes corresponding to different anchors. Instead, anchor-free methods is lighter, getting rid of redundant hyper-parameters, but gains few attention. In this paper, we propose the first purely anchor-free temporal localization method, which is both efficient and effective. Our model includes (i) an end-to-end trainable basic predictor, (ii) a saliency-based refinement module to gather more valuable boundary features for each proposal with a novel boundary pooling, and (iii) several consistency constraints to make sure our model can find the accurate boundary given arbitrary proposals. Extensive experiments show that our method beats all anchor-based and actionness-guided methods with a remarkable margin on THUMOS14, achieving state-of-the-art results, and comparable ones on ActivityNet v1.3. Code is available at https://github.com/TencentYoutuResearch/ActionDetection-AFSD., Accepted by CVPR2021

Cardiac tagging magnetic resonance imaging (t-MRI) is the gold standard for regional myocardium deformation and cardiac strain estimation. However, this technique has not been widely used in clinical diagnosis, as a result of the difficulty of motion tracking encountered with t-MRI images. In this paper, we propose a novel deep learning-based fully unsupervised method for in vivo motion tracking on t-MRI images. We first estimate the motion field (INF) between any two consecutive t-MRI frames by a bi-directional generative diffeomorphic registration neural network. Using this result, we then estimate the Lagrangian motion field between the reference frame and any other frame through a differentiable composition layer. By utilizing temporal information to perform reasonable estimations on spatio-temporal motion fields, this novel method provides a useful solution for motion tracking and image registration in dynamic medical imaging. Our method has been validated on a representative clinical t-MRI dataset; the experimental results show that our method is superior to conventional motion tracking methods in terms of landmark tracking accuracy and inference efficiency., Comment: CVPR 2021 Oral

Analysis of bispectral difference plays a critical role in various applications that involve rays propagating in a light absorbing medium. In general, the bispectral difference is obtained by subtracting signals at two individual wave-lengths captured by ordinary digital cameras, which tends to inherit the drawbacks of conventional cameras in dynamic range, response speed and quantization precision. In this paper, we propose a novel method to obtain a bispectral difference image using an event camera with temporally modulated illumination. Our method is rooted in a key observation on the analogy between the bispectral photometry principle of the participating medium and the event generating mechanism in an event camera. By carefully modulating the bispectral illumination, our method allows to read out the bispectral difference directly from triggered events. Experiments using a prototype imaging system have verified the feasibility of this novel usage of event cameras in photometry based vision tasks, such as 3D shape reconstruction in water.

Aligning two partially-overlapped 3D line reconstructions in Euclidean space is challenging, as we need to simultaneously solve correspondences and relative pose between line reconstructions. This paper proposes a neural network based method and it has three modules connected in sequence: (i) a Multilayer Perceptron (MLP) based network takes Plucker representations of lines as inputs, to extract discriminative line-wise features and matchabilities (how likely each line is going to have a match), (ii) an Optimal Transport (OT) layer takes two-view line-wise features and matchabilities as inputs to estimate a 2D joint probability matrix, with each item describes the matchness of a line pair, and (iii) line pairs with Top-K matching probabilities are fed to a 2-line minimal solver in a RANSAC framework to estimate a six Degree-of-Freedom (6-DoF) rigid transformation. Experiments on both indoor and outdoor datasets show that registration (rotation and translation) precision of our method outperforms baselines significantly.