Showing papers in "International Journal of Computer Vision in 2020"

Deep Learning for Generic Object Detection: A Survey

Abstract: Object detection, one of the most fundamental and challenging problems in computer vision, seeks to locate object instances from a large number of predefined categories in natural images. Deep learning techniques have emerged as a powerful strategy for learning feature representations directly from data and have led to remarkable breakthroughs in the field of generic object detection. Given this period of rapid evolution, the goal of this paper is to provide a comprehensive survey of the recent achievements in this field brought about by deep learning techniques. More than 300 research contributions are included in this survey, covering many aspects of generic object detection: detection frameworks, object feature representation, object proposal generation, context modeling, training strategies, and evaluation metrics. We finish the survey by identifying promising directions for future research.

Grad-CAM: Visual Explanations from Deep Networks via Gradient-Based Localization

Abstract: We propose a technique for producing ‘visual explanations’ for decisions from a large class of Convolutional Neural Network (CNN)-based models, making them more transparent and explainable. Our approach—Gradient-weighted Class Activation Mapping (Grad-CAM), uses the gradients of any target concept (say ‘dog’ in a classification network or a sequence of words in captioning network) flowing into the final convolutional layer to produce a coarse localization map highlighting the important regions in the image for predicting the concept. Unlike previous approaches, Grad-CAM is applicable to a wide variety of CNN model-families: (1) CNNs with fully-connected layers (e.g.VGG), (2) CNNs used for structured outputs (e.g.captioning), (3) CNNs used in tasks with multi-modal inputs (e.g.visual question answering) or reinforcement learning, all without architectural changes or re-training. We combine Grad-CAM with existing fine-grained visualizations to create a high-resolution class-discriminative visualization, Guided Grad-CAM, and apply it to image classification, image captioning, and visual question answering (VQA) models, including ResNet-based architectures. In the context of image classification models, our visualizations (a) lend insights into failure modes of these models (showing that seemingly unreasonable predictions have reasonable explanations), (b) outperform previous methods on the ILSVRC-15 weakly-supervised localization task, (c) are robust to adversarial perturbations, (d) are more faithful to the underlying model, and (e) help achieve model generalization by identifying dataset bias. For image captioning and VQA, our visualizations show that even non-attention based models learn to localize discriminative regions of input image. We devise a way to identify important neurons through Grad-CAM and combine it with neuron names (Bau et al. in Computer vision and pattern recognition, 2017) to provide textual explanations for model decisions. Finally, we design and conduct human studies to measure if Grad-CAM explanations help users establish appropriate trust in predictions from deep networks and show that Grad-CAM helps untrained users successfully discern a ‘stronger’ deep network from a ‘weaker’ one even when both make identical predictions. Our code is available at https://github.com/ramprs/grad-cam/, along with a demo on CloudCV (Agrawal et al., in: Mobile cloud visual media computing, pp 265–290. Springer, 2015) (http://gradcam.cloudcv.org) and a video at http://youtu.be/COjUB9Izk6E.

Diverse Image-to-Image Translation via Disentangled Representations

Abstract: Image-to-image translation aims to learn the mapping between two visual domains. There are two main challenges for many applications: (1) the lack of aligned training pairs and (2) multiple possible outputs from a single input image. In this work, we present an approach based on disentangled representation for producing diverse outputs without paired training images. To achieve diversity, we propose to embed images onto two spaces: a domain-invariant content space capturing shared information across domains and a domain-specific attribute space. Using the disentangled features as inputs greatly reduces mode collapse. To handle unpaired training data, we introduce a novel cross-cycle consistency loss. Qualitative results show that our model can generate diverse and realistic images on a wide range of tasks. We validate the effectiveness of our approach through extensive evaluation.

CornerNet: Detecting Objects as Paired Keypoints

Abstract: We propose CornerNet, a new approach to object detection where we detect an object bounding box as a pair of keypoints, the top-left corner and the bottom-right corner, using a single convolution neural network. By detecting objects as paired keypoints, we eliminate the need for designing a set of anchor boxes commonly used in prior single-stage detectors. In addition to our novel formulation, we introduce corner pooling, a new type of pooling layer that helps the network better localize corners. Experiments show that CornerNet achieves a 42.1% AP on MS COCO, outperforming all existing one-stage detectors.

The Open Images Dataset V4: Unified Image Classification, Object Detection, and Visual Relationship Detection at Scale

Abstract: We present Open Images V4, a dataset of 9.2M images with unified annotations for image classification, object detection and visual relationship detection. The images have a Creative Commons Attribution license that allows to share and adapt the material, and they have been collected from Flickr without a predefined list of class names or tags, leading to natural class statistics and avoiding an initial design bias. Open Images V4 offers large scale across several dimensions: 30.1M image-level labels for 19.8k concepts, 15.4M bounding boxes for 600 object classes, and 375k visual relationship annotations involving 57 classes. For object detection in particular, we provide $$15\times $$ more bounding boxes than the next largest datasets (15.4M boxes on 1.9M images). The images often show complex scenes with several objects (8 annotated objects per image on average). We annotated visual relationships between them, which support visual relationship detection, an emerging task that requires structured reasoning. We provide in-depth comprehensive statistics about the dataset, we validate the quality of the annotations, we study how the performance of several modern models evolves with increasing amounts of training data, and we demonstrate two applications made possible by having unified annotations of multiple types coexisting in the same images. We hope that the scale, quality, and variety of Open Images V4 will foster further research and innovation even beyond the areas of image classification, object detection, and visual relationship detection.

Cognitive Mapping and Planning for Visual Navigation

Abstract: We introduce a neural architecture for navigation in novel environments. Our proposed architecture learns to map from first-person views and plans a sequence of actions towards goals in the environment. The Cognitive Mapper and Planner (CMP) is based on two key ideas: (a) a unified joint architecture for mapping and planning, such that the mapping is driven by the needs of the task, and (b) a spatial memory with the ability to plan given an incomplete set of observations about the world. CMP constructs a top-down belief map of the world and applies a differentiable neural net planner to produce the next action at each time step. The accumulated belief of the world enables the agent to track visited regions of the environment. We train and test CMP on navigation problems in simulation environments derived from scans of real world buildings. Our experiments demonstrate that CMP outperforms alternate learning-based architectures, as well as, classical mapping and path planning approaches in many cases. Furthermore, it naturally extends to semantically specified goals, such as “going to a chair”. We also deploy CMP on physical robots in indoor environments, where it achieves reasonable performance, even though it is trained entirely in simulation.

Loss-Sensitive Generative Adversarial Networks on Lipschitz Densities

Abstract: In this paper, we present the Lipschitz regularization theory and algorithms for a novel Loss-Sensitive Generative Adversarial Network (LS-GAN). Specifically, it trains a loss function to distinguish between real and fake samples by designated margins, while learning a generator alternately to produce realistic samples by minimizing their losses. The LS-GAN further regularizes its loss function with a Lipschitz regularity condition on the density of real data, yielding a regularized model that can better generalize to produce new data from a reasonable number of training examples than the classic GAN. We will further present a Generalized LS-GAN (GLS-GAN) and show it contains a large family of regularized GAN models, including both LS-GAN and Wasserstein GAN, as its special cases. Compared with the other GAN models, we will conduct experiments to show both LS-GAN and GLS-GAN exhibit competitive ability in generating new images in terms of the Minimum Reconstruction Error (MRE) assessed on a separate test set. We further extend the LS-GAN to a conditional form for supervised and semi-supervised learning problems, and demonstrate its outstanding performance on image classification tasks.

Single Image Dehazing via Multi-scale Convolutional Neural Networks with Holistic Edges

Abstract: Single image dehazing has been a challenging problem which aims to recover clear images from hazy ones. The performance of existing image dehazing methods is limited by hand-designed features and priors. In this paper, we propose a multi-scale deep neural network for single image dehazing by learning the mapping between hazy images and their transmission maps. The proposed algorithm consists of a coarse-scale net which predicts a holistic transmission map based on the entire image, and a fine-scale net which refines dehazed results locally. To train the multi-scale deep network, we synthesize a dataset comprised of hazy images and corresponding transmission maps based on the NYU Depth dataset. In addition, we propose a holistic edge guided network to refine edges of the estimated transmission map. Extensive experiments demonstrate that the proposed algorithm performs favorably against the state-of-the-art methods on both synthetic and real-world images in terms of quality and speed.

GMS: Grid-Based Motion Statistics for Fast, Ultra-robust Feature Correspondence

Abstract: Feature matching aims at generating correspondences across images, which is widely used in many computer vision tasks. Although considerable progress has been made on feature descriptors and fast matching for initial correspondence hypotheses, selecting good ones from them is still challenging and critical to the overall performance. More importantly, existing methods often take a long computational time, limiting their use in real-time applications. This paper attempts to separate true correspondences from false ones at high speed. We term the proposed method (GMS) grid-based motion Statistics, which incorporates the smoothness constraint into a statistic framework for separation and uses a grid-based implementation for fast calculation. GMS is robust to various challenging image changes, involving in viewpoint, scale, and rotation. It is also fast, e.g., take only 1 or 2 ms in a single CPU thread, even when 50K correspondences are processed. This has important implications for real-time applications. What’s more, we show that incorporating GMS into the classic feature matching and epipolar geometry estimation pipeline can significantly boost the overall performance. Finally, we integrate GMS into the well-known ORB-SLAM system for monocular initialization, resulting in a significant improvement.

Self-Supervised Model Adaptation for Multimodal Semantic Segmentation

Abstract: Learning to reliably perceive and understand the scene is an integral enabler for robots to operate in the real-world. This problem is inherently challenging due to the multitude of object types as well as appearance changes caused by varying illumination and weather conditions. Leveraging complementary modalities can enable learning of semantically richer representations that are resilient to such perturbations. Despite the tremendous progress in recent years, most multimodal convolutional neural network approaches directly concatenate feature maps from individual modality streams rendering the model incapable of focusing only on the relevant complementary information for fusion. To address this limitation, we propose a mutimodal semantic segmentation framework that dynamically adapts the fusion of modality-specific features while being sensitive to the object category, spatial location and scene context in a self-supervised manner. Specifically, we propose an architecture consisting of two modality-specific encoder streams that fuse intermediate encoder representations into a single decoder using our proposed self-supervised model adaptation fusion mechanism which optimally combines complementary features. As intermediate representations are not aligned across modalities, we introduce an attention scheme for better correlation. In addition, we propose a computationally efficient unimodal segmentation architecture termed AdapNet++ that incorporates a new encoder with multiscale residual units and an efficient atrous spatial pyramid pooling that has a larger effective receptive field with more than $$10\,\times $$ fewer parameters, complemented with a strong decoder with a multi-resolution supervision scheme that recovers high-resolution details. Comprehensive empirical evaluations on Cityscapes, Synthia, SUN RGB-D, ScanNet and Freiburg Forest benchmarks demonstrate that both our unimodal and multimodal architectures achieve state-of-the-art performance while simultaneously being efficient in terms of parameters and inference time as well as demonstrating substantial robustness in adverse perceptual conditions.

Realistic Speech-Driven Facial Animation with GANs

Abstract: Speech-driven facial animation is the process that automatically synthesizes talking characters based on speech signals. The majority of work in this domain creates a mapping from audio features to visual features. This approach often requires post-processing using computer graphics techniques to produce realistic albeit subject dependent results. We present an end-to-end system that generates videos of a talking head, using only a still image of a person and an audio clip containing speech, without relying on handcrafted intermediate features. Our method generates videos which have (a) lip movements that are in sync with the audio and (b) natural facial expressions such as blinks and eyebrow movements. Our temporal GAN uses 3 discriminators focused on achieving detailed frames, audio-visual synchronization, and realistic expressions. We quantify the contribution of each component in our model using an ablation study and we provide insights into the latent representation of the model. The generated videos are evaluated based on sharpness, reconstruction quality, lip-reading accuracy, synchronization as well as their ability to generate natural blinks.

Temporal Action Detection with Structured Segment Networks

Abstract: This paper addresses an important and challenging task, namely detecting the temporal intervals of actions in untrimmed videos. Specifically, we present a framework called structured segment network (SSN). It is built on temporal proposals of actions. SSN models the temporal structure of each action instance via a structured temporal pyramid. On top of the pyramid, we further introduce a decomposed discriminative model comprising two classifiers, respectively for classifying actions and determining completeness. This allows the framework to effectively distinguish positive proposals from background or incomplete ones, thus leading to both accurate recognition and precise localization. These components are integrated into a unified network that can be efficiently trained in an end-to-end manner. Additionally, a simple yet effective temporal action proposal scheme, dubbed temporal actionness grouping is devised to generate high quality action proposals. We further study the importance of the decomposed discriminative model and discover a way to achieve similar accuracy using a single classifier, which is also complementary with the original SSN design. On two challenging benchmarks, THUMOS’14 and ActivityNet, our method remarkably outperforms previous state-of-the-art methods, demonstrating superior accuracy and strong adaptivity in handling actions with various temporal structures.

Weakly-supervised Semantic Guided Hashing for Social Image Retrieval

Abstract: Hashing has been widely investigated for large-scale image retrieval due to its search effectiveness and computation efficiency. In this work, we propose a novel Semantic Guided Hashing method coupled with binary matrix factorization to perform more effective nearest neighbor image search by simultaneously exploring the weakly-supervised rich community-contributed information and the underlying data structures. To uncover the underlying semantic information from the weakly-supervised user-provided tags, the binary matrix factorization model is leveraged for learning the binary features of images while the problem of imperfect tags is well addressed. The uncovered semantic information enables to well guide the discrete hash code learning. The underlying data structures are discovered by adaptively learning a discriminative data graph, which makes the learned hash codes preserve the meaningful neighbors. To the best of our knowledge, the proposed method is the first work that incorporates the hash code learning, the semantic information mining and the data structure discovering into one unified framework. Besides, the proposed method is extended to one deep approach for the optimal compatibility of discriminative feature learning and hash code learning. Experiments are conducted on two widely-used social image datasets and the proposed method achieves encouraging performance compared with the state-of-the-art hashing methods.

Text2Sign: Towards Sign Language Production Using Neural Machine Translation and Generative Adversarial Networks.

Abstract: We present a novel approach to automatic Sign Language Production using recent developments in Neural Machine Translation (NMT), Generative Adversarial Networks, and motion generation. Our system is capable of producing sign videos from spoken language sentences. Contrary to current approaches that are dependent on heavily annotated data, our approach requires minimal gloss and skeletal level annotations for training. We achieve this by breaking down the task into dedicated sub-processes. We first translate spoken language sentences into sign pose sequences by combining an NMT network with a Motion Graph. The resulting pose information is then used to condition a generative model that produces photo realistic sign language video sequences. This is the first approach to continuous sign video generation that does not use a classical graphical avatar. We evaluate the translation abilities of our approach on the PHOENIX14T Sign Language Translation dataset. We set a baseline for text-to-gloss translation, reporting a BLEU-4 score of 16.34/15.26 on dev/test sets. We further demonstrate the video generation capabilities of our approach for both multi-signer and high-definition settings qualitatively and quantitatively using broadcast quality assessment metrics.

Convolutional Networks with Adaptive Inference Graphs

Abstract: Do convolutional networks really need a fixed feed-forward structure? What if, after identifying the high-level concept of an image, a network could move directly to a layer that can distinguish fine-grained differences? Currently, a network would first need to execute sometimes hundreds of intermediate layers that specialize in unrelated aspects. Ideally, the more a network already knows about an image, the better it should be at deciding which layer to compute next. In this work, we propose convolutional networks with adaptive inference graphs (ConvNet-AIG) that adaptively define their network topology conditioned on the input image. Following a high-level structure similar to residual networks (ResNets), ConvNet-AIG decides for each input image on the fly which layers are needed. In experiments on ImageNet we show that ConvNet-AIG learns distinct inference graphs for different categories. Both ConvNet-AIG with 50 and 101 layers outperform their ResNet counterpart, while using $$20\%$$ and $$38\%$$ less computations respectively. By grouping parameters into layers for related classes and only executing relevant layers, ConvNet-AIG improves both efficiency and overall classification quality. Lastly, we also study the effect of adaptive inference graphs on the susceptibility towards adversarial examples. We observe that ConvNet-AIG shows a higher robustness than ResNets, complementing other known defense mechanisms.

Deep Image Prior

Abstract: Deep convolutional networks have become a popular tool for image generation and restoration. Generally, their excellent performance is imputed to their ability to learn realistic image priors from a large number of example images. In this paper, we show that, on the contrary, the structure of a generator network is sufficient to capture a great deal of low-level image statistics prior to any learning. In order to do so, we show that a randomly-initialized neural network can be used as a handcrafted prior with excellent results in standard inverse problems such as denoising, super-resolution, and inpainting. Furthermore, the same prior can be used to invert deep neural representations to diagnose them, and to restore images based on flash-no flash input pairs. Apart from its diverse applications, our approach highlights the inductive bias captured by standard generator network architectures. It also bridges the gap between two very popular families of image restoration methods: learning-based methods using deep convolutional networks and learning-free methods based on handcrafted image priors such as self-similarity (Code and supplementary material are available at https://dmitryulyanov.github.io/deep_image_prior).

Augmented Autoencoders: Implicit 3D Orientation Learning for 6D Object Detection

Abstract: We propose a real-time RGB-based pipeline for object detection and 6D pose estimation. Our novel 3D orientation estimation is based on a variant of the Denoising Autoencoder that is trained on simulated views of a 3D model using Domain Randomization. This so-called Augmented Autoencoder has several advantages over existing methods: It does not require real, pose-annotated training data, generalizes to various test sensors and inherently handles object and view symmetries. Instead of learning an explicit mapping from input images to object poses, it provides an implicit representation of object orientations defined by samples in a latent space. Our pipeline achieves state-of-the-art performance on the T-LESS dataset both in the RGB and RGB-D domain. We also evaluate on the LineMOD dataset where we can compete with other synthetically trained approaches. We further increase performance by correcting 3D orientation estimates to account for perspective errors when the object deviates from the image center and show extended results. Our code is available here https://github.com/DLR-RM/AugmentedAutoencoder.

EdgeStereo: An Effective Multi-task Learning Network for Stereo Matching and Edge Detection

Abstract: Recently, leveraging on the development of end-to-end convolutional neural networks, deep stereo matching networks have achieved remarkable performance far exceeding traditional approaches. However, state-of-the-art stereo frameworks still have difficulties at finding correct correspondences in texture-less regions, detailed structures, small objects and near boundaries, which could be alleviated by geometric clues such as edge contours and corresponding constraints. To improve the quality of disparity estimates in these challenging areas, we propose an effective multi-task learning network, EdgeStereo, composed of a disparity estimation branch and an edge detection branch, which enables end-to-end predictions of both disparity map and edge map. To effectively incorporate edge cues, we propose the edge-aware smoothness loss and edge feature embedding for inter-task interactions. It is demonstrated that based on our unified model, edge detection task and stereo matching task can promote each other. In addition, we design a compact module called residual pyramid to replace the commonly-used multi-stage cascaded structures or 3-D convolution based regularization modules in current stereo matching networks. By the time of the paper submission, EdgeStereo achieves state-of-art performance on the FlyingThings3D dataset, KITTI 2012 and KITTI 2015 stereo benchmarks, outperforming other published stereo matching methods by a noteworthy margin. EdgeStereo also achieves comparable generalization performance for disparity estimation because of the incorporation of edge cues.

Learning Single-Image 3D Reconstruction by Generative Modelling of Shape, Pose and Shading

Abstract: We present a unified framework tackling two problems: class-specific 3D reconstruction from a single image, and generation of new 3D shape samples. These tasks have received considerable attention recently; however, most existing approaches rely on 3D supervision, annotation of 2D images with keypoints or poses, and/or training with multiple views of each object instance. Our framework is very general: it can be trained in similar settings to existing approaches, while also supporting weaker supervision. Importantly, it can be trained purely from 2D images, without pose annotations, and with only a single view per instance. We employ meshes as an output representation, instead of voxels used in most prior work. This allows us to reason over lighting parameters and exploit shading information during training, which previous 2D-supervised methods cannot. Thus, our method can learn to generate and reconstruct concave object classes. We evaluate our approach in various settings, showing that: (i) it learns to disentangle shape from pose and lighting; (ii) using shading in the loss improves performance compared to just silhouettes; (iii) when using a standard single white light, our model outperforms state-of-the-art 2D-supervised methods, both with and without pose supervision, thanks to exploiting shading cues; (iv) performance improves further when using multiple coloured lights, even approaching that of state-of-the-art 3D-supervised methods; (v) shapes produced by our model capture smooth surfaces and fine details better than voxel-based approaches; and (vi) our approach supports concave classes such as bathtubs and sofas, which methods based on silhouettes cannot learn.

Modeling Human Motion with Quaternion-Based Neural Networks

Abstract: Previous work on predicting or generating 3D human pose sequences regresses either joint rotations or joint positions. The former strategy is prone to error accumulation along the kinematic chain, as well as discontinuities when using Euler angles or exponential maps as parameterizations. The latter requires re-projection onto skeleton constraints to avoid bone stretching and invalid configurations. This work addresses both limitations. QuaterNet represents rotations with quaternions and our loss function performs forward kinematics on a skeleton to penalize absolute position errors instead of angle errors. We investigate both recurrent and convolutional architectures and evaluate on short-term prediction and long-term generation. For the latter, our approach is qualitatively judged as realistic as recent neural strategies from the graphics literature. Our experiments compare quaternions to Euler angles as well as exponential maps and show that only a very short context is required to make reliable future predictions. Finally, we show that the standard evaluation protocol for Human3.6M produces high variance results and we propose a simple solution.

Curriculum Model Adaptation with Synthetic and Real Data for Semantic Foggy Scene Understanding

Abstract: This work addresses the problem of semantic scene understanding under fog. Although marked progress has been made in semantic scene understanding, it is mainly concentrated on clear-weather scenes. Extending semantic segmentation methods to adverse weather conditions such as fog is crucial for outdoor applications. In this paper, we propose a novel method, named Curriculum Model Adaptation (CMAda), which gradually adapts a semantic segmentation model from light synthetic fog to dense real fog in multiple steps, using both labeled synthetic foggy data and unlabeled real foggy data. The method is based on the fact that the results of semantic segmentation in moderately adverse conditions (light fog) can be bootstrapped to solve the same problem in highly adverse conditions (dense fog). CMAda is extensible to other adverse conditions and provides a new paradigm for learning with synthetic data and unlabeled real data. In addition, we present four other main stand-alone contributions: (1) a novel method to add synthetic fog to real, clear-weather scenes using semantic input; (2) a new fog density estimator; (3) a novel fog densification method for real foggy scenes without known depth; and (4) the Foggy Zurich dataset comprising 3808 real foggy images, with pixel-level semantic annotations for 40 images with dense fog. Our experiments show that (1) our fog simulation and fog density estimator outperform their state-of-the-art counterparts with respect to the task of semantic foggy scene understanding (SFSU); (2) CMAda improves the performance of state-of-the-art models for SFSU significantly, benefiting both from our synthetic and real foggy data. The foggy datasets and code are publicly available.

GANimation: One-Shot Anatomically Consistent Facial Animation

Abstract: Recent advances in generative adversarial networks (GANs) have shown impressive results for the task of facial expression synthesis. The most successful architecture is StarGAN (Choi et al. in CVPR, 2018), that conditions GANs’ generation process with images of a specific domain, namely a set of images of people sharing the same expression. While effective, this approach can only generate a discrete number of expressions, determined by the content and granularity of the dataset. To address this limitation, in this paper, we introduce a novel GAN conditioning scheme based on action units (AU) annotations, which describes in a continuous manifold the anatomical facial movements defining a human expression. Our approach allows controlling the magnitude of activation of each AU and combining several of them. Additionally, we propose a weakly supervised strategy to train the model, that only requires images annotated with their activated AUs, and exploit a novel self-learned attention mechanism that makes our network robust to changing backgrounds, lighting conditions and occlusions. Extensive evaluation shows that our approach goes beyond competing conditional generators both in the capability to synthesize a much wider range of expressions ruled by anatomically feasible muscle movements, as in the capacity of dealing with images in the wild. The code of this work is publicly available at https://github.com/albertpumarola/GANimation.

The Unmanned Aerial Vehicle Benchmark: Object Detection, Tracking and Baseline

Abstract: With the increasing popularity of Unmanned Aerial Vehicles (UAVs) in computer vision-related applications, intelligent UAV video analysis has recently attracted the attention of an increasing number of researchers. To facilitate research in the UAV field, this paper presents a UAV dataset with 100 videos featuring approximately 2700 vehicles recorded under unconstrained conditions and 840k manually annotated bounding boxes. These UAV videos were recorded in complex real-world scenarios and pose significant new challenges, such as complex scenes, high density, small objects, and large camera motion, to the existing object detection and tracking methods. These challenges have encouraged us to define a benchmark for three fundamental computer vision tasks, namely, object detection, single object tracking (SOT) and multiple object tracking (MOT), on our UAV dataset. Specifically, our UAV benchmark facilitates evaluation and detailed analysis of state-of-the-art detection and tracking methods on the proposed UAV dataset. Furthermore, we propose a novel approach based on the so-called Context-aware Multi-task Siamese Network (CMSN) model that explores new cues in UAV videos by judging the consistency degree between objects and contexts and that can be used for SOT and MOT. The experimental results demonstrate that our model could make tracking results more robust in both SOT and MOT, showing that the current tracking and detection methods have limitations in dealing with the proposed UAV benchmark and that further research is indeed needed.

EKLT: Asynchronous Photometric Feature Tracking Using Events and Frames

Abstract: We present a method that leverages the complementarity of event cameras and standard cameras to track visual features with low-latency. Event cameras are novel sensors that output pixel-level brightness changes, called “events”. They offer significant advantages over standard cameras, namely a very high dynamic range, no motion blur, and a latency in the order of microseconds. However, because the same scene pattern can produce different events depending on the motion direction, establishing event correspondences across time is challenging. By contrast, standard cameras provide intensity measurements (frames) that do not depend on motion direction. Our method extracts features on frames and subsequently tracks them asynchronously using events, thereby exploiting the best of both types of data: the frames provide a photometric representation that does not depend on motion direction and the events provide low-latency updates. In contrast to previous works, which are based on heuristics, this is the first principled method that uses raw intensity measurements directly, based on a generative event model within a maximum-likelihood framework. As a result, our method produces feature tracks that are both more accurate (subpixel accuracy) and longer than the state of the art, across a wide variety of scenes.

Tensorized Multi-view Subspace Representation Learning

Abstract: Self-representation based subspace learning has shown its effectiveness in many applications. In this paper, we promote the traditional subspace representation learning by simultaneously taking advantages of multiple views and prior constraint. Accordingly, we establish a novel algorithm termed as Tensorized Multi-view Subspace Representation Learning. To exploit different views, the subspace representation matrices of different views are regarded as a low-rank tensor, which effectively models the high-order correlations of multi-view data. To incorporate prior information, a constraint matrix is devised to guide the subspace representation learning within a unified framework. The subspace representation tensor equipped with a low-rank constraint models elegantly the complementary information among different views, reduces redundancy of subspace representations, and then improves the accuracy of subsequent tasks. We formulate the model with a tensor nuclear norm minimization problem constrained with $$\ell _{2,1}$$ -norm and linear equalities. The minimization problem is efficiently solved by using an Augmented Lagrangian Alternating Direction Minimization method. Extensive experimental results on diverse multi-view datasets demonstrate the effectiveness of our algorithm.

RGB-IR Person Re-identification by Cross-Modality Similarity Preservation

Abstract: Person re-identification (Re-ID) is an important problem in video surveillance for matching pedestrian images across non-overlapping camera views. Currently, most works focus on RGB-based Re-ID. However, RGB images are not well suited to a dark environment; consequently, infrared (IR) imaging becomes necessary for indoor scenes with low lighting and 24-h outdoor scene surveillance systems. In such scenarios, matching needs to be performed between RGB images and IR images, which exhibit different visual characteristics; this cross-modality matching problem is more challenging than RGB-based Re-ID due to the lack of visible colour information in IR images. To address this challenge, we study the RGB-IR cross-modality Re-ID (RGB-IR Re-ID) problem. Rather than applying existing cross-modality matching models that operate under the assumption of identical data distributions between training and testing sets to handle the discrepancy between RGB and IR modalities for Re-ID, we cast learning shared knowledge for cross-modality matching as the problem of cross-modality similarity preservation. We exploit same-modality similarity as the constraint to guide the learning of cross-modality similarity along with the alleviation of modality-specific information, and finally propose a Focal Modality-Aware Similarity-Preserving Loss. To further assist the feature extractor in extracting shared knowledge, we design a modality-gated node as a universal representation of both modality-specific and shared structures for constructing a structure-learnable feature extractor called Modality-Gated Extractor. For validation, we construct a new multi-modality Re-ID dataset, called SYSU-MM01, to enable wider study of this problem. Extensive experiments on this SYSU-MM01 dataset show the effectiveness of our method. Download link of dataset: https://github.com/wuancong/SYSU-MM01.

Learning 3D Shape Completion Under Weak Supervision

Abstract: We address the problem of 3D shape completion from sparse and noisy point clouds, a fundamental problem in computer vision and robotics. Recent approaches are either data-driven or learning-based: Data-driven approaches rely on a shape model whose parameters are optimized to fit the observations; Learning-based approaches, in contrast, avoid the expensive optimization step by learning to directly predict complete shapes from incomplete observations in a fully-supervised setting. However, full supervision is often not available in practice. In this work, we propose a weakly-supervised learning-based approach to 3D shape completion which neither requires slow optimization nor direct supervision. While we also learn a shape prior on synthetic data, we amortize, i.e., learn, maximum likelihood fitting using deep neural networks resulting in efficient shape completion without sacrificing accuracy. On synthetic benchmarks based on ShapeNet (Chang et al. Shapenet: an information-rich 3d model repository, 2015. arXiv:1512.03012 ) and ModelNet (Wu et al., in: Proceedings of IEEE conference on computer vision and pattern recognition (CVPR), 2015) as well as on real robotics data from KITTI (Geiger et al., in: Proceedings of IEEE conference on computer vision and pattern recognition (CVPR), 2012) and Kinect (Yang et al., 3d object dense reconstruction from a single depth view, 2018. arXiv:1802.00411 ), we demonstrate that the proposed amortized maximum likelihood approach is able to compete with the fully supervised baseline of Dai et al. (in: Proceedings of IEEE conference on computer vision and pattern recognition (CVPR), 2017) and outperforms the data-driven approach of Engelmann et al. (in: Proceedings of the German conference on pattern recognition (GCPR), 2016), while requiring less supervision and being significantly faster.

Robust Attentional Aggregation of Deep Feature Sets for Multi-view 3D Reconstruction

Abstract: We study the problem of recovering an underlying 3D shape from a set of images. Existing learning based approaches usually resort to recurrent neural nets, e.g., GRU, or intuitive pooling operations, e.g., max/mean poolings, to fuse multiple deep features encoded from input images. However, GRU based approaches are unable to consistently estimate 3D shapes given different permutations of the same set of input images as the recurrent unit is permutation variant. It is also unlikely to refine the 3D shape given more images due to the long-term memory loss of GRU. Commonly used pooling approaches are limited to capturing partial information, e.g., max/mean values, ignoring other valuable features. In this paper, we present a new feed-forward neural module, named AttSets, together with a dedicated training algorithm, named FASet, to attentively aggregate an arbitrarily sized deep feature set for multi-view 3D reconstruction. The AttSets module is permutation invariant, computationally efficient and flexible to implement, while the FASet algorithm enables the AttSets based network to be remarkably robust and generalize to an arbitrary number of input images. We thoroughly evaluate FASet and the properties of AttSets on multiple large public datasets. Extensive experiments show that AttSets together with FASet algorithm significantly outperforms existing aggregation approaches.

Statistical Modeling of Craniofacial Shape and Texture

Abstract: We present a fully-automatic statistical 3D shape modeling approach and apply it to a large dataset of 3D images, the Headspace dataset, thus generating the first public shape-and-texture 3D morphable model (3DMM) of the full human head. Our approach is the first to employ a template that adapts to the dataset subject before dense morphing. This is fully automatic and achieved using 2D facial landmarking, projection to 3D shape, and mesh editing. In dense template morphing, we improve on the well-known Coherent Point Drift algorithm, by incorporating iterative data-sampling and alignment. Our evaluations demonstrate that our method has better performance in correspondence accuracy and modeling ability when compared with other competing algorithms. We propose a texture map refinement scheme to build high quality texture maps and texture model. We present several applications that include the first clinical use of craniofacial 3DMMs in the assessment of different types of surgical intervention applied to a craniosynostosis patient group.

Train Sparsely, Generate Densely: Memory-Efficient Unsupervised Training of High-Resolution Temporal GAN

Abstract: Training of generative adversarial network (GAN) on a video dataset is a challenge because of the sheer size of the dataset and the complexity of each observation. In general, the computational cost of training GAN scales exponentially with the resolution. In this study, we present a novel memory efficient method of unsupervised learning of high-resolution video dataset whose computational cost scales only linearly with the resolution. We achieve this by designing the generator model as a stack of small sub-generators and training the model in a specific way. We train each sub-generator with its own specific discriminator. At the time of the training, we introduce between each pair of consecutive sub-generators an auxiliary subsampling layer that reduces the frame-rate by a certain ratio. This procedure can allow each sub-generator to learn the distribution of the video at different levels of resolution. We also need only a few GPUs to train a highly complex generator that far outperforms the predecessor in terms of inception scores.