ECCV 2020 Papers

Deep learning (DL) methods have revolutionized the paradigm of computer vision tasks and DL-based video compression is becoming a hot topic. This paper proposes a deep video compression method to simultaneously encode multiple frames with Frame-Conv3D and differential modulation. We first adopt Frame-Conv3D instead of traditional Channel-Conv3D for efficient multi-frame fusion. When generating the binary representation, the multi-frame differential modulation is utilized to alleviate the effect of quantization noise. By analyzing the forward and backward computing flow of the modulator, we identify that this technique can make full use of past frames' information to remove the redundancy between multiple frames, and thus achieves better performance. A dropout scheme combined with the differential modulator is proposed to enable bit rate optimization within a single model. Experimental results show that the proposed approach outperforms the H.264 and H.265 codecs in the region of low bit rate. Compared with recent DL-based methods, our model also achieves competitive performance. "

Existing image inpainting methods often produce artifacts when dealing with large holes in real applications. To address this challenge, we propose an iterative inpainting method with a feedback mechanism. Specifically, we introduce a deep generative model which not only outputs an inpainting result but also a corresponding confidence map. Using this map as feedback, it progressively fills the hole by trusting only high-confidence pixels inside the hole at each iteration and focuses on the remaining pixels in the next iteration. As it reuses partial predictions from the previous iterations as known pixels, this process gradually improves the result. In addition, we propose a guided upsampling network to enable generation of high-resolution inpainting results. We achieve this by extending the Contextual Attention module to borrow high-resolution feature patches in the input image. Furthermore, to mimic real object removal scenarios, we collect a large object mask dataset and synthesize more realistic training data that better simulates user inputs. Experiments show that our method significantly outperforms existing methods in both quantitative and qualitative evaluations. "

Learning disentangled representation of data without supervision is an important step towards improving the interpretability of generative models. Despite recent advances in disentangled representation learning, existing approaches often suffer from the trade-off between representation learning and generation performance (i.e. improving generation quality sacrifices disentanglement performance). We propose an Information-Distillation Generative Adversarial Network (ID-GAN), a simple yet generic framework that can easily incorporate the existing state-of-the-art models for both disentanglement learning and high-fidelity synthesis. Our method learns disentangled representation using VAE-based models, and distills the learned representation together with additional nuisance variable to the separate GAN-based generator for high-fidelity synthesis. To ensure that both generative models are aligned to render the same generative factors, we additionally constrain the GAN generator to maximize the mutual information between the learned latent code and the output. Despite the simplicity, we show that the proposed method is surprisingly effective, achieving comparable image generation quality to the state-of-the-arts using the disentangled representation. We also show that the proposed decomposition leads to efficient and stable model design, and demonstrate it on high-resolution image synthesis task (1024x1024 pixels) for the first time using the disentangled representations."

Generating photorealistic images of human faces at scale remains a prohibitively difficult task using computer graphics approaches. This is because these require the simulation of light to be photorealistic, which in turn requires physically accurate modelling of geometry, materials, and light sources, for both the head and the surrounding scene. Non-photorealistic renders however are increasingly easy to produce. In contrast to computer graphics approaches, generative models learned from more readily available 2D image data have been shown to produce samples of human faces that are hard to distinguish from real data. The process of learning usually corresponds to a loss of control over the shape and appearance of the generated images. For instance, even simple disentangling tasks such as modifying the hair independently of the face, which is trivial to accomplish in a computer graphics approach, remains an open research question. In this work, we propose an algorithm that matches a non-photorealistic, synthetically generated image to a latent vector of a pretrained StyleGAN2 model which, in turn, maps the vector to a photorealistic image of a person of the same pose, expression, hair, and lighting. In contrast to most previous work, we require no synthetic training data. To the best of our knowledge, this is the first algorithm of its kind to work at a resolution of 1K and represents a significant leap forward in visual realism."

Temporal Activity Localization via Language (TALL) in video is a recently proposed challenging vision task, and tackling it requires fine-grained understanding of the video content, however, this is overlooked by most of the existing works. In this paper, we propose a novel TALL method which builds a Hierarchical Visual-Textual Graph to model interactions between the objects and words as well as among the objects to jointly understand the video contents and the language. We also design a convolutional network with cross-channel communication mechanism to further encourage the information passing between the visual and textual modalities. Finally, we propose a loss function that enforces alignment of the visual representation of the localized activity and the sentence representation, so that the model can predict more accurate temporal boundaries. We evaluated our proposed method on two popular benchmark datasets: Charades-STA and ActivityNet Captions, and achieved state-of-the-art performances on both datasets. Code is available at \url{https://github.com/forwchen/HVTG}."

Generating diverse and natural behaviors is one of the long-standing goals for creating intelligent characters in the animated world. In this paper, we propose an unsupervised method for generating long-range, diverse and plausible behaviors to achieve a specific goal location. Our proposed method learns to model the motion of human by decomposing a long-range generation task in a hierarchical manner. Given the starting and ending states, a memory bank is used to retrieve motion references as source material for short-range clip generation. We first propose to explicitly disentangle the provided motion material into style and content counterparts via bi-linear transformation modelling, where diverse synthesis is achieved by free-form combination of these two components. The short-range clips are then connected to form a long-range motion sequence. Without ground truth annotation, we propose a parameterized bi-directional interpolation scheme to guarantee the physical validity and visual naturalness of generated results. On large-scale skeleton dataset, we show that the proposed method is able to synthesise long-range, diverse and plausible motion, which is also generalizable to unseen motion data during testing. Moreover, we demonstrate the generated sequences are useful as subgoals for actual physical execution in the animated world. Please refer to our project page~ootnote{https://sites.google.com/view/hsnms/home} for more synthesised results."

We consider the problem of estimating a parametric model of 3D human mesh from a single image. While there has been substantial recent progress in this area with direct regression of model parameters, these methods only implicitly exploit the human body kinematic structure, leading to sub-optimal use of the model prior. In this work, we address this gap by proposing a new technique for regression of human parametric model that is explicitly informed by the known hierarchical structure, including joint interdependencies of the model. This results in a strong prior-informed design of the regressor architecture and an associated hierarchical optimization that is flexible to be used in conjunction with the current standard frameworks for 3D human mesh recovery. We demonstrate these aspects by means of extensive experiments on standard benchmark datasets, showing how our proposed new design outperforms several existing and popular methods, establishing new state-of-the-art results. By considering joint interdependencies, our method is equipped to infer joints even under data corruptions, which we demonstrate by conducting experiments under varying degrees of occlusion."

Current age datasets lie in a long-tailed distribution, which brings difficulties to describe the aging mechanism for the imbalance ages. To alleviate it, we design a novel facial age prior to guide the aging mechanism modeling. To explore the age effects on facial images, we propose a Disentangled Adversarial Autoencoder (DAAE) to disentangle the facial images into three independent factors: age, identity and extraneous information. To avoid the ""wash away"" of age and identity information in face aging process, we propose a hierarchical conditional generator by passing the disentangled identity and age embeddings to the high-level and low-level layers with class-conditional BatchNorm. Finally, a disentangled adversarial learning mechanism is introduced to boost the image quality for face aging. In this way, when manipulating the age distribution, DAAE can achieve face aging with arbitrary ages. Further, given an input face image, the mean value of the learned age posterior distribution can be treated as an age estimator. These indicate that DAAE can efficiently and accurately estimate the age distribution by a disentangling manner. DAAE is the first attempt to achieve facial age analysis tasks, including face aging with arbitrary ages, exemplar-based face aging and age estimation, in an universal framework. The qualitative and quantitative experiments demonstrate the superiority of DAAE on five popular datasets, including CACD2000, Morph, UTKFace, FG-NET and AgeDB."

The main purpose of RGB-D salient object detection (SOD) is how to better integrate and utilize cross-modal fusion information. In this paper, we explore these issues from a new perspective. We integrate the features of different modalities through densely connected structures and use their mixed features to generate dynamic filters with receptive fields of different sizes. In the end, we implement a kind of more flexible and efficient multi-scale cross-modal feature processing, i.e. dynamic dilated pyramid module. In order to make the predictions have sharper edges and consistent saliency regions, we design a hybrid enhanced loss function to further optimize the results. This loss function is also validated to be effective in the single-modal RGB SOD task. In terms of six metrics, the proposed method outperforms the existing twelve methods on eight challenging benchmark datasets. A large number of experiments verify the effectiveness of the proposed module and loss function. Our code, model and results are available at \url{https://github.com/lartpang/HDFNet}."

State-of-the-art two-stage object detectors apply a classifier to a sparse set of object proposals, relying on region-wise features extracted by RoIPool or RoIAlign as inputs. The region-wise features, in spite of aligning well with the proposal locations, may still lack the crucial context information which is necessary for filtering out noisy background detections, as well as recognizing objects possessing no distinctive appearances. To address this issue, we present a simple but effective Hierarchical Context Embedding (HCE) framework, which can be applied as a plug-and-play component, to facilitate the classification ability of a series of region-based detectors by mining contextual cues. Specifically, to advance the recognition of context-dependent object categories, we propose an image-level categorical embedding module which leverages the holistic image-level context to learn object-level concepts. Then, novel RoI features are generated by exploiting hierarchically embedded context information beneath both whole images and interested regions, which are also complementary to conventional RoI features. Moreover, to make full use of our hierarchical contextual RoI features, we propose the early-and-late fusion strategies (\ie, feature fusion and confidence fusion), which can be combined to boost the classification accuracy of region-based detectors. Comprehensive experiments demonstrate that our HCE framework is flexible and generalizable, leading to significant and consistent improvements upon various region-based detectors, including FPN, Cascade R-CNN and Mask R-CNN."

Predicting where people can walk in a scene is important for many tasks, including autonomous driving systems and human behavior analysis. Yet learning a computational model for this purpose is challenging due to semantic ambiguity and a lack of labeled data: current datasets only have labels on where people $ extit{are}$, not where they $ extit{could be}$. We tackle this problem by leveraging information from existing datasets, without any additional labeling. We first augment the set of valid walkable regions by propagating person observations between images, utilizing 3D information and temporal coherence, leading to $ extit{Hidden Footprints}$. We then design a training strategy that combines a class-balanced classification loss with a contextual adversarial loss to learn from sparse observations, thus obtaining a model that predicts a walkability map of a given scene. We evaluate our model on the Waymo and Cityscapes datasets, demonstrating superior performance against baselines and state-of-the-art models."

In this paper, we present a Haze-Aware Representation Distillation Generative Adversarial Network named HardGAN for single-image dehazing. Unlike previous studies that intend to model the transmission map and global atmospheric light jointly to restore a clear image, we solve this regression problem by a multi-scale structure neural network embedded with our proposed Haze-Aware Representation Distillation (HARD) layer. Moreover, we re-introduce to utilize the normalization layer skillfully instead of stacking with the convolution layer directly as before to avoid the useful information wash away, as claimed in many image quality enhancement studies. Extensive experiment on several synthetic benchmark datasets as well as the NTIRE 2020 real-world images show our proposed multi-layer GAN-based network with HARD performs favorably against the state-of-the-art methods in terms of PSNR, SSIM, LPIPS, and human subjective evaluation."

but useful","Triplet loss is an extremely common approach to distance metric learning. Representations of images from the same class are optimized to be mapped closer together in an embedding space than representations of images from different classes. Much work on triplet losses focuses on selecting the most useful triplets of images to consider, with strategies that select dissimilar examples from the same class or similar examples from different classes. The consensus of previous research is that optimizing with the extit{hardest} negative examples leads to bad training behavior. That's a problem -- these hardest negatives are literally the cases where the distance metric fails to capture semantic similarity. In this paper, we characterize the space of triplets and derive why hard negatives make triplet loss training fail. We offer a simple fix to the loss function and show that, with this fix, optimizing with hard negative examples becomes feasible. This leads to more generalizable features, and image retrieval results that outperform state of the art for datasets with high intra-class variance."

Local feature matching is a critical part of many computer vision pipelines, including among others Structure-from-Motion, SLAM, and Visual Localization. However, due to limitations in the descriptors, raw matches are often contaminated by a majority of outliers. As a result, outlier detection is a fundamental problem in computer vision and a wide range of approaches, from simple checks based on descriptor similarity to geometric verification, have been proposed over the last decades. In recent years, deep learning-based approaches to outlier detection have become popular. Unfortunately, the corresponding works rarely compare with strong classical baselines. In this paper we revisit handcrafted approaches to outlier filtering. Based on best practices, we propose a hierarchical pipeline for effective outlier detection as well as integrate novel ideas which in sum lead to an efficient and competitive approach to outlier rejection. We show that our approach, although not relying on learning, is more than competitive to both recent learned works as well as handcrafted approaches, both in terms of efficiency and effectiveness. The code is available at https://github.com/cavalli1234/AdaLAM}{https://github.com/cavalli1234/AdaLAM."

3D hand pose estimation is still far from a well-solved problem mainly due to the highly nonlinear dynamics of hand pose and the difficulties of modeling its inherent structural dependencies. To address this issue, we connect this structured output learning problem with the structured modeling framework in sequence transduction field. Standard transduction models like Transformer adopt an autoregressive connection to capture dependencies from previously generated tokens and further correlate this information with the input sequence in order to prioritize the set of relevant input tokens for current token generation. To borrow wisdom from this structured learning framework while avoiding the sequential modeling for hand pose, taking a 3D point set as input, we propose to leverage the Transformer architecture with a novel non-autoregressive structured decoding mechanism. Specifically, instead of using previously generated results, our decoder utilizes a reference hand pose to provide equivalent dependencies among hand joints for each output joint generation. By imposing the reference structural dependencies, we can correlate the information with the input 3D points through a multi-head attention mechanism, aiming to discover informative points from different perspectives, towards each hand joint localization. We demonstrate our model's effectiveness over multiple challenging hand pose datasets, comparing with several state-of-the-art methods."

We revisit the classical problem of 3D shape interpolation and propose a novel, physically plausible approach based on Hamiltonian dynamics. While most prior work focuses on synthetic input shapes, our formulation is designed to be applicable to real-world scans with imperfect input correspondences and various types of noise. To that end, we use recent progress on dynamic thin shell simulation and divergence-free shape deformation and combine them to address the inverse problem of finding a plausible intermediate sequence for two input shapes. In comparison to prior work that mainly focuses on small distortion of consecutive frames, we explicitly model volume preservation and momentum conservation, as well as an anisotropic local distortion model. We argue that, in order to get a robust interpolation for imperfect inputs, we need to model the input noise explicitly which results in an alignment based formulation. Finally, we show a qualitative and quantitative improvement over prior work on a broad range of synthetic and scanned data. Besides being more robust to noisy inputs, our method yields exactly volume preserving intermediate shapes, avoids self-intersections and is scalable to high resolution scans."

In this paper, we investigate a new visual restoration task, termed as hallucinating visual instances in total absentia (HVITA). Unlike conventional image inpainting task that works on images with only part of a visual instance missing, HVITA concerns scenarios where an object is completely absent from the scene. This seemingly minor difference in fact makes the HVITA a much challenging task, as the restoration algorithm would have to not only infer the category of the object in total absentia, but also hallucinate an object of which the appearance is consistent with the background. Towards solving HVITA, we propose an end-to-end deep approach that explicitly looks into the global semantics within the image. Specifically, we transform the input image to a semantic graph, wherein each node corresponds to a detected object in the scene. We then adopt a Graph Convolutional Network on top of the scene graph to estimate the category of the missing object in the masked region, and finally introduce a Generative Adversarial Module to carry out the hallucination. Experiments on COCO, Visual Genome and NYU Depth v2 datasets demonstrate that the proposed approach yields truly encouraging and visually plausible results."

3D hand reconstruction from images is a widely-studied problem in computer vision and graphics, and has a particularly high relevance for virtual and augmented reality. Although several 3D hand reconstruction approaches leverage hand models as a strong prior to resolve ambiguities and achieve more robust results, most existing models account only for the hand shape and poses and do not model the texture. To fill this gap, in this work we present HTML, the first parametric texture model of human hands. Our model spans several dimensions of hand appearance variability (e.g. related to gender, ethnicity, or age) and only requires a commodity camera for data acquisition. Experimentally, we demonstrate that our appearance model can be used to tackle a range of challenging problems such as 3D hand reconstruction from a single monocular image. Furthermore, our appearance model can be used to define a neural rendering layer that enables training with a self-supervised photometric loss. We make our model publicly available."

Recent work in network quantization produced state-of-the-art results using mixed precision quantization. An imperative requirement for many efficient edge device hardware implementations is that their quantizers are uniform and with power-of-two thresholds. In this work, we introduce the Hardware Friendly Mixed Precision Quantization Block (HMQ) in order to meet this requirement. The HMQ is a mixed precision quantization block that repurposes the Gumbel-Softmax estimator into a smooth estimator of a pair of quantization parameters, namely, bit-width and threshold. HMQs use this to search over a finite space of quantization schemes. Empirically, we apply HMQs to quantize classification models trained on CIFAR10 and ImageNet. For ImageNet, we quantize four different architectures and show that, in spite of the added restrictions to our quantization scheme, we achieve competitive and, in some cases, state-of-the-art results."