ECCV 2020 Papers

Video-based person re-identification aims to match pedestrians with the consecutive video sequences. While a rich line of work focuses solely on extracting the motion features from pedestrian videos, we show in this paper that the temporal coherence plays a more critical role. To distill the temporal coherence part of video representation from frame representations, we propose a simple yet effective Adversarial Feature Augmentation (AFA) method, which highlights the temporal coherence features by introducing adversarial augmented temporal motion noise. Specifically, we disentangle the video representation into the temporal coherence and motion parts and randomly change the scale of the temporal motion features as the adversarial noise. The proposed AFA method is a general lightweight component that can be readily incorporated into various methods with negligible cost. We conduct extensive experiments on three challenging datasets including MARS, iLIDS-VID, and DukeMTMC-VideoReID, and the experimental results verify our argument and demonstrate the effectiveness of the proposed method."

Future prediction, especially in long-range videos, requires reasoning from current and past observations. In this work, we address questions of temporal extent, scaling, and level of semantic abstraction with a flexible multi-granular temporal aggregation framework. We show that it is possible to achieve state of the art in both next action and dense anticipation with simple techniques such as max-pooling and attention. To demonstrate the anticipation capabilities of our model, we conduct experiments on Breakfast, 50Salads, and EPIC-Kitchens datasets, where we achieve state-of-the-art results. With minimal modifications, our model can also be extended for video segmentation and action recognition."

The connection between visual input and tactile sensing is critical for object manipulation tasks such as grasping, pushing, and maneuvering. In this work, we introduce the challenging task of estimating a set of tactile physical properties from visual information. We aim to build a model that learns the complex mapping between visual information and tactile physical properties. We construct a first of its kind image-tactile dataset with over 400 multiview image sequences and the corresponding tactile properties. A total of 15 tactile physical properties across categories including friction, compliance, adhesion, texture, and thermal conductance are measured and then estimated by our models. We develop a cross-modal framework comprised of an adversarial objective and a novel visuo-tactile cluster classification loss. Additionally, we develop a neural architecture search framework capable of learning to select optimal combinations of viewing angles for estimating a given physical property."

Pedestrian detection is a core problem in computer vision that sees broad application in video surveillance and, more recently, in advanced driving assistance systems. Despite its broad application and interest, it remains a challenging problem in part due to the vast range of conditions under which it must be robust. Pedestrian detection at nighttime and during adverse weather conditions is particularly challenging, which is one of the reasons why thermal and multispectral approaches have been become popular in recent years. In this paper, we propose a novel approach to domain adaptation that significantly improves pedestrian detection performance in the thermal domain. The key idea behind our technique is to adapt an RGB-trained detection network to simultaneously solve two related tasks. An auxiliary classification task that distinguishes between daytime and nighttime thermal images is added to the main detection task during domain adaptation. The internal representation learned to perform this classification task is used to condition a YOLOv3 detector at multiple points in order to improve its adaptation to the thermal domain. We validate the effectiveness of task-conditioned domain adaptation by comparing with the state-of-the-art on the KAIST Multispectral Pedestrian Detection Benchmark. To the best of our knowledge, our proposed task-conditioned approach achieves the best single-modality detection results."

We propose to learn a deep neural network for JPEG image compression, which predicts image-specific optimized quantization tables fully compatible with the standard JPEG encoder and decoder. Moreover, our approach provides the capability to learn task-specific quantization tables in a principled way by adjusting the objective function of the network. The main challenge to realize this idea is that there exist non-differentiable components in the encoder such as run-length encoding and Huffman coding and it is not straightforward to predict the probability distribution of the quantized image representations. We address these issues by learning a differentiable loss function that approximates bitrates using simple network blocks---two MLPs and an LSTM. We evaluate the proposed algorithm using multiple task-specific losses---two for semantic image understanding and another two for conventional image compression---and demonstrate the effectiveness of our approach to the individual tasks. "

The deep hashing based retrieval method is widely adopted in large-scale image and video retrieval. However, there is little investigation on its security. In this paper, we propose a novel method, dubbed deep hashing targeted attack (DHTA), to study the targeted attack on such retrieval. Specifically, we first formulate the targeted attack as a point-to-set optimization, which minimizes the average distance between the hash code of an adversarial example and those of a set of objects with the target label. Then we design a novel component-voting scheme to obtain an anchor code as the representative of the set of hash codes of objects with the target label, whose optimality guarantee is also theoretically derived. To balance the performance and perceptibility, we propose to minimize the Hamming distance between the hash code of the adversarial example and the anchor code under the $ll^\infty$ restriction on the perturbation. Extensive experiments verify that DHTA is effective in attacking both deep hashing based image retrieval and video retrieval."

When people deliver a speech, they naturally move heads, and this rhythmic head motion conveys linguistic information. However, generating a lip-synced video while moving head naturally is challenging. While remarkably successful, existing works either generate still talking-face videos or rely on landmark/video frames as sparse/dense mapping guidance to generate head movements, which leads to unrealistic or uncontrollable video synthesis. To overcome the limitations, we propose a 3D-aware generative network along with a hybrid embedding module and a non-linear composition module. Through modeling the head motion and facial expressions explicitly, manipulating 3D animation carefully, and embedding reference images dynamically, our approach achieves controllable, photorealistic, and temporally coherent talking-head videos with natural head movements. Thoughtful experiments on several standard benchmarks demonstrate that our method achieves significantly better results than the state-of-the-art methods in both quantitative and qualitative comparisons. The code is available on https://github.com/lelechen63/Talking-head-Generation-with-Rhythmic-Head-Motion"

We present an autoencoder-based semi-supervised approach to classify perceived human emotions from walking styles obtained from videos or motion-captured data and represented as sequences of 3D poses. Given the motion on each joint in the pose at each time step extracted from 3D pose sequences, we hierarchically pool these joint motions in a bottom-up manner in the encoder, following the kinematic chains in the human body. We also constrain the latent embeddings of the encoder to contain the space of psychologically-motivated affective features underlying the gaits. We train the decoder to reconstruct the motions per joint per time step in a top-down manner from the latent embeddings. For the annotated data, we also train a classifier to map the latent embeddings to emotion labels. Our semi-supervised approach achieves a mean average precision of 0.84 on the Emotion-Gait benchmark dataset, which contains both labeled and unlabeled gaits collected from multiple sources. We outperform current state-of-art algorithms for both emotion recognition and action recognition from 3D gaits by 7%-23% on the absolute. More importantly, we improve the average precision by 10%-50% on the absolute on classes that each makes up less than 25% of the labeled part of the Emotion-Gait benchmark dataset."

Tables are information-rich structured objects in document images. While significant work has been done in localizing tables as graphic objects in document images, only limited attempts exist on table structure recognition. Most existing literature on structure recognition depends on extraction of meta-features from the {\sc pdf} document or on the optical character recognition (OCR) models to extract low-level layout features from the image. However, these methods fail to generalize well because of the absence of meta-features or errors made by the OCR when there is a significant variance in table layouts and text organization. In our work, we focus on tables that have complex structures, dense content, and varying layouts with no dependency on meta-features and/or OCR.

We introduce TV show Retrieval (TVR), a new multimodal retrieval dataset. TVR requires systems to understand both videos and their associated subtitle (dialogue) texts, making it more realistic. The dataset contains 109K queries collected on 21.8K videos from 6 TV shows of diverse genres, where each query is associated with a tight temporal window. The queries are also labeled with query types that indicate whether each of them is more related to video or subtitle or both, allowing for in-depth analysis of the dataset and the methods that built on top of it. Strict qualification and post-annotation verification tests are applied to ensure the quality of the collected data. Additionally, we present several baselines and a novel Cross-modal Moment Localization (XML) network for multimodal moment retrieval tasks. The proposed XML model uses a late fusion design with a novel Convolutional StartEnd detector (ConvSE), surpassing baselines by a large margin and with better efficiency, providing a strong starting point for future work."

We introduce a simple and versatile framework for image-to-image translation. We unearth the importance of normalization layers, and provide a carefully designed two-stream generative model with newly proposed feature transformations in a coarse-to-fine fashion. This allows multi-scale semantic structure information and style representation to be effectively captured and fused by the network, permitting our method to scale to various tasks in both unsupervised and supervised settings. No additional constraints (e.g., cycle consistency) are needed, contributing to a very clean and simple method. Multi-modal image synthesis with arbitrary style control is made possible. A systematic study compares the proposed method with several state-of-the-art task-specific baselines, verifying its effectiveness in both perceptual quality and quantitative evaluations."

Compositional visual question answering requires reasoning over both semantic and geometry object relations. We propose a novel tiered reasoning method that dynamically selects object level candidates based on language representations and generates robust pairwise relations within the selected candidate objects. The proposed tiered relation reasoning method can be compatible with the majority of the existing visual reasoning frameworks, leading to significant performance improvement with very little extra computational cost. Moreover, we propose a policy network that decides the appropriate reasoning steps based on question complexity and current reasoning status. In experiments, our model achieves state-of-the-art performance on two VQA datasets. "

During training, the weights of a Deep Neural Network (DNN) are optimized from a random initialization towards a nearly optimum value minimizing a loss function. Only this final state of the weights is typically kept for testing, while the wealth of information on the geometry of the weight space, accumulated over the descent towards the minimum is discarded. In this work we propose to make use of this knowledge and leverage it for computing the distributions of the weights of the DNN. This can be further used for estimating the epistemic uncertainty of the DNN by aggregating predictions from an ensemble of networks sampled from these distributions. To this end we introduce a method for tracking the trajectory of the weights during optimization, that does neither require any change in the architecture, nor in the training procedure. We evaluate our method, TRADI, on standard classification and regression benchmarks, and on out-of-distribution detection for classification and semantic segmentation. We achieve competitive results, while preserving computational efficiency in comparison to ensemble approaches."

Multimodal sentiment analysis (MSA) has been widely investigated in both computer vision and natural language processing. However, studies on the imperfect data especially with missing values are still far from success and challenging, even though such an issue is ubiquitous in the real world. Although previous works show the promising performance by exploiting the low-rank structures of the fused features, only the first-order statistics of the temporal dynamics are concerned. To this end, we propose a novel network architecture termed Time Product Fusion Network (TPFN), which takes the high-order statistics over both modalities and temporal dynamics into account. We construct the fused features by the outer product along adjacent time-steps, such that richer modal and temporal interactions are utilized. In addition, we claim that the low-rank structures can be obtained by regularizing the Frobenius norm of latent factors instead of the fused features. Experiments on CMU-MOSI and CMU-MOSEI datasets show that TPFN can compete with state-of-the art approaches in multimodal sentiment analysis in cases of both random and structured missing values."

This paper proposes a novel deep convolutional model, Tri-Points Based Line Segment Detector (TP-LSD), to detect line segments in an image at real-time speed. The previous related methods typically use the two-step strategy, relying on either heuristic post-process or extra classifier. To realize one-step detection with a faster and more compact model, we introduce the tri-points representation, converting the line segment detection to the end-to-end prediction of a root-point and two endpoints for each line segment. TP-LSD has two branches: tri-points extraction branch and line segmentation branch. The former predicts the heat map of root-points and the two displacement maps of endpoints. The latter segments the pixels on straight lines out from background. Moreover, the line segmentation map is reused in the first branch as structural prior. We propose an additional novel evaluation metric and evaluate our method on Wireframe and YorkUrban datasets, demonstrating not only the competitive accuracy compared to the most recent methods, but also the real-time run speed up to extbf{78 FPS} with the 320 × 320 input."

We introduce TIDE, a framework and associated toolbox for analyzing the sources of error in object detection and instance segmentation algorithms. Importantly, our framework is applicable across datasets and can be applied directly to output prediction files without required knowledge of the underlying prediction system. Thus, our framework can be used as a drop-in replacement for the standard mAP computation while providing a comprehensive analysis of each model's strengths and weaknesses. We segment errors into six types and, crucially, are the first to introduce a technique for measuring the contribution of each error in a way that isolates its effect on overall performance. We show that such a representation is critical for drawing accurate, comprehensive conclusions through in-depth analysis across 4 datasets and 7 recognition models."

With the flourish of differentiable neural architecture search (NAS), automatically searching latency-constrained architectures gives a new perspective to reduce human labor and expertise. However, the searched architectures are usually suboptimal in accuracy and may have large jitters around the target latency. In this paper, we rethink three freedoms of differentiable NAS, i.e. operation-level, depth-level and width-level, and propose a novel method, named Three-Freedom NAS (TF-NAS), to achieve both good classification accuracy and precise latency constraint. For the operation-level, we present a bi-sampling search algorithm to moderate the operation collapse. For the depth-level, we introduce a sink-connecting search space to ensure the mutual exclusion between skip and other candidate operations, as well as eliminate the architecture redundancy. For the width-level, we propose an elasticity-scaling strategy that achieves precise latency constraint in a progressively fine-grained manner. Experiments on ImageNet demonstrate the effectiveness of TF-NAS. Particularly, our searched TF-NAS-A obtains 76.9% top-1 accuracy, achieving state-of-the-art results with less latency. Code is available at https://github.com/AberHu/TF-NAS."

In this paper, we propose a simple yet effective approach, named Triple Excitation Network, to reinforce the training of video salient object detection (VSOD) from three aspects, spatial, temporal, and online excitations. These excitation mechanisms are designed following the spirit of curriculum learning and aim to reduce learning ambiguities at the beginning of training by selectively exciting feature activations using ground truth. Then we gradually reduce the weight of ground truth excitations by a curriculum rate and replace it by a curriculum complementary map for better and faster convergence. In particular, the spatial excitation strengthens feature activations for clear object boundaries, while the temporal excitation imposes motions to emphasize spatio-temporal salient regions. Spatial and temporal excitations can combat the saliency shifting problem and conflict between spatial and temporal features of VSOD. Furthermore, our semi-curriculum learning design enables the first online refinement strategy for VSOD, which allows exciting and boosting saliency responses during testing without re-training. The proposed triple excitations can easily plug in different VSOD methods. Extensive experiments show the effectiveness of all three excitation methods and the proposed method outperforms state-of-the-art image and video salient object detection methods."

Fusing data from multiple modalities provides more information to train machine learning systems. However, it is prohibitively expensive and time-consuming to label each modality with a large amount of data, which leads to a crucial problem of such semi-supervised multi-modal learning. Existing methods suffer from either ineffective fusion across modalities or lack of theoretical results under proper assumptions. In this paper, we propose a novel information-theoretic approach \-- namely, extbf{T}otal extbf{C}orrelation extbf{G}ain extbf{M}aximization (TCGM) \--- for semi-supervised multi-modal learning, which is endowed with promising properties: (i) it can utilize effectively the information across different modalities of unlabeled data points to facilitate training classifiers of each modality (ii) has theoretical guarantee to have theoretical guarantee to identify Bayesian classifiers, i.e., the ground truth posteriors of all modalities. Specifically, by maximizing TC-induced loss (namely TC gain) over classifiers of all modalities, these classifiers can cooperatively discover the equivalent class of ground-truth classifiers; and identify the unique ones by leveraging a limited percentage of labeled data. We apply our method and can achieve state-of-the-art results on various datasets, including the Newsgroup dataset, Emotion recognition (IEMOCAP and MOSI) and Medical Imaging (Alzheimer’s Disease Neuroimaging Initiative).