Pix2Shape: Towards Unsupervised Learning of 3D Scenes from Images using a View-based Representation
Université de Montreal（蒙特利尔大学）；Element AI；Algolux；McGill University（麦吉尔大学）
Sai Rajeswar, Fahim Mannan, Florian Golemo, Jérôme Parent-Lévesque, David Vazquez, Derek Nowrouzezahrai, Aaron Courville
备注：
This is a pre-print of an article published in International Journal of Computer Vision. The final authenticated version is available online at: https://doi.org/10.1007/s11263-020-01322-1
发表信息：
International Journal of Computer Vision, (2020), 1-16
We infer and generate three-dimensional (3D) scene information from a single input image and without supervision. This problem is under-explored, with most prior work relying on supervision from, e.g., 3D ground-truth, multiple images of a scene, image silhouettes or key-points. We propose Pix2Shape, an approach to solve this problem with four components: (i) an encoder that infers the latent 3D representation from an image, (ii) a decoder that generates an explicit 2.5D surfel-based reconstruction of a scene from the latent code (iii) a differentiable renderer that synthesizes a 2D image from the surfel representation, and (iv) a critic network trained to discriminate between images generated by the decoder-renderer and those from a training distribution. Pix2Shape can generate complex 3D scenes that scale with the view-dependent on-screen resolution, unlike representations that capture world-space resolution, i.e., voxels or meshes. We show that Pix2Shape learns a consistent scene representation in its encoded latent space and that the decoder can then be applied to this latent representation in order to synthesize the scene from a novel viewpoint. We evaluate Pix2Shape with experiments on the ShapeNet dataset as well as on a novel benchmark we developed, called 3D-IQTT, to evaluate models based on their ability to enable 3d spatial reasoning. Qualitative and quantitative evaluation demonstrate Pix2Shape's ability to solve scene reconstruction, generation, and understanding tasks.
Tübingen AI Center（图宾根人工智能中心）；University of Tübingen（图宾根大学）；EPFL（洛桑联邦理工学院）；Max Planck Institute for Intelligent Systems（马克斯·普朗克智能系统研究所）；University of Cambridge（剑桥大学）；Amazon（亚马逊）
Unsupervised image-to-image translation methods have received a lot of attention in the last few years. Multiple techniques emerged tackling the initial challenge from different perspectives. Some focus on learning as much as possible from several target style images for translations while other make use of object detection in order to produce more realistic results on content-rich scenes. In this work, we assess how a method that has initially been developed for single object translation performs on more diverse and content-rich images. Our work is based on the FUNIT[1] framework and we train…
Shanghai Key Laboratory of Intelligent Information Processing, School of Computer Science, Fudan University（复旦大学计算机科学技术学院上海市智能信息处理重点实验室）
Visual scenes are extremely rich in diversity, not only because there are infinite combinations of objects and background, but also because the observations of the same scene may vary greatly with the change of viewpoints. When observing a visual scene that contains multiple objects from multiple viewpoints, humans are able to perceive the scene in a compositional way from each viewpoint, while achieving the so-called "object constancy" across different viewpoints, even though the exact viewpoints are untold. This ability is essential for humans to identify the same object while moving and to …
This work develops a novel end-to-end deep unsupervised learning method based on convolutional neural network (CNN) with pseudo-classes for remote sensing scene representation. First, we introduce center points as the centers of the pseudo classes and the training samples can be allocated with pseudo labels based on the center points. Therefore, the CNN model, which is used to extract features from the scenes, can be trained supervised with the pseudo labels. Moreover, a pseudo-center loss is developed to decrease the variance between the samples and the corresponding pseudo center point. The …
Huazhong University of Science and Technology（华中科技大学）；Huawei International Co. Ltd.（华为国际有限公司）
Optical flow has achieved great success under clean scenes, but suffers from restricted performance under foggy scenes. To bridge the clean-to-foggy domain gap, the existing methods typically adopt the domain adaptation to transfer the motion knowledge from clean to synthetic foggy domain. However, these methods unexpectedly neglect the synthetic-to-real domain gap, and thus are erroneous when applied to real-world scenes. To handle the practical optical flow under real foggy scenes, in this work, we propose a novel unsupervised cumulative domain adaptation optical flow (UCDA-Flow) framework: …
There has been a recent surge in methods that aim to decompose and segment scenes into multiple objects in an unsupervised manner, i.e., unsupervised multi-object segmentation. Performing such a task is a long-standing goal of computer vision, offering to unlock object-level reasoning without requiring dense annotations to train segmentation models. Despite significant progress, current models are developed and trained on visually simple scenes depicting mono-colored objects on plain backgrounds. The natural world, however, is visually complex with confounding aspects such as diverse textures …
Beijing Institute for General Artificial Intelligence（北京通用人工智能研究院）；ShanghaiTech University（上海科技大学）
Understanding realistic visual scene images together with language descriptions is a fundamental task towards generic visual understanding. Previous works have shown compelling comprehensive results by building hierarchical structures for visual scenes (e.g., scene graphs) and natural languages (e.g., dependency trees), individually. However, how to construct a joint vision-language (VL) structure has barely been investigated. More challenging but worthwhile, we introduce a new task that targets on inducing such a joint VL structure in an unsupervised manner. Our goal is to bridge the visual s…
Centre for Artificial Intelligence, University of Technology Sydney（悉尼科技大学人工智能中心）
We consider the unsupervised scene adaptation problem of learning from both labeled source data and unlabeled target data. Existing methods focus on minoring the inter-domain gap between the source and target domains. However, the intra-domain knowledge and inherent uncertainty learned by the network are under-explored. In this paper, we propose an orthogonal method, called memory regularization in vivo to exploit the intra-domain knowledge and regularize the model training. Specifically, we refer to the segmentation model itself as the memory module, and minor the discrepancy of the two class…
Zhejiang University（浙江大学）；Tencent（腾讯）；University of Hong Kong（香港大学）
Deep learning-based scene text detection can achieve preferable performance, powered with sufficient labeled training data. However, manual labeling is time consuming and laborious. At the extreme, the corresponding annotated data are unavailable. Exploiting synthetic data is a very promising solution except for domain distribution mismatches between synthetic datasets and real datasets. To address the severe domain distribution mismatch, we propose a synthetic-to-real domain adaptation method for scene text detection, which transfers knowledge from synthetic data (source domain) to real data …
Multi-view clustering methods have been a focus in recent years because of their superiority in clustering performance. However, typical traditional multi-view clustering algorithms still have shortcomings in some aspects, such as removal of redundant information, utilization of various views and fusion of multi-view features. In view of these problems, this paper proposes a new multi-view clustering method, low-rank subspace multi-view clustering based on adaptive graph regularization. We construct two new data matrix decomposition models into a unified optimization model. In this framework, …
Unsupervised Collaborative Domain Adaptation for Driving Scene Parsing
College of Electronic and Information Engineering, Tongji University（同济大学电子与信息学院）；Department of Control Science and Engineering, Harbin Institute of Technology（控制科学与工程系，哈尔滨工业大学）；Department of Vehicle Control System and Software Development, NIO（车辆控制系统与软件开发部，蔚来汽车）；School of Computer Science and Technology, Tongji University（计算机科学与技术学院，同济大学）；Key Laboratory of Embedded System and Service Computing (Ministry of Education), Tongji University（嵌入式系统与服务计算重点实验室（教育部），同济大学）
National Institute of Informatics（国立信息学研究所）；Wuhan University（武汉大学）；National Tsing Hua University（国立清华大学）
Understanding foggy image sequence in the driving scenes is critical for autonomous driving, but it remains a challenging task due to the difficulty in collecting and annotating real-world images of adverse weather. Recently, the self-training strategy has been considered a powerful solution for unsupervised domain adaptation, which iteratively adapts the model from the source domain to the target domain by generating target pseudo labels and re-training the model. However, the selection of confident pseudo labels inevitably suffers from the conflict between sparsity and accuracy, both of whic…
KAIST（韩国科学技术院）；MIT CSAIL（麻省理工学院计算机科学与人工智能实验室）；University of California, Merced（加州大学默塞德分校）
Visual events are usually accompanied by sounds in our daily lives. We pose the question: Can the machine learn the correspondence between visual scene and the sound, and localize the sound source only by observing sound and visual scene pairs like human? In this paper, we propose a novel unsupervised algorithm to address the problem of localizing the sound source in visual scenes. A two-stream network structure which handles each modality, with attention mechanism is developed for sound source localization. Moreover, although our network is formulated within the unsupervised learning framewor…
Tianjin University（天津大学）；University of Wollongong（伍伦贡大学）；Santa Clara University（圣塔克拉拉大学）
View synthesis aims to generate novel views from one or more given source views. Although existing methods have achieved promising performance, they usually require paired views of different poses to learn a pixel transformation. This paper proposes an unsupervised network to learn such a pixel transformation from a single source viewpoint. In particular, the network consists of a token transformation module (TTM) that facilities the transformation of the features extracted from a source viewpoint image into an intrinsic representation with respect to a pre-defined reference pose and a view ge…
Simon Fraser University（西蒙弗雷泽大学）；Microsoft Research（微软研究院）
We present an unsupervised approach for factorizing object appearance into highlight, shading, and albedo layers, trained by multi-view real images. To do so, we construct a multi-view dataset by collecting numerous customer product photos online, which exhibit large illumination variations that make them suitable for training of reflectance separation and can facilitate object-level decomposition. The main contribution of our approach is a proposed image representation based on local color distributions that allows training to be insensitive to the local misalignments of multi-view images. In…
A classical problem in computer vision is to infer a 3D scene representation from few images that can be used to render novel views at interactive rates. Previous work focuses on reconstructing pre-defined 3D representations, e.g. textured meshes, or implicit representations, e.g. radiance fields, and often requires input images with precise camera poses and long processing times for each novel scene. In this work, we propose the Scene Representation Transformer (SRT), a method which processes posed or unposed RGB images of a new area, infers a "set-latent scene representation", and synthesise…
Multi-View People Detection in Large Scenes via Supervised View-Wise Contribution Weighting
College of Computer Science and Software Engineering, Shenzhen University（深圳大学计算机与软件学院）；Guangdong Laboratory of Artificial Intelligence and Digital Economy (SZ)（人工智能与数字经济广东省实验室（深圳））；Department of Computer Science, City University of Hong Kong（香港城市大学计算机科学系）
Beihang University（北京航空航天大学）；Tencent（腾讯）；University of Amsterdam（阿姆斯特丹大学）；Peng Cheng Laboratory（鹏城实验室）
Intrinsic image decomposition, which is an essential task in computer vision, aims to infer the reflectance and shading of the scene. It is challenging since it needs to separate one image into two components. To tackle this, conventional methods introduce various priors to constrain the solution, yet with limited performance. Meanwhile, the problem is typically solved by supervised learning methods, which is actually not an ideal solution since obtaining ground truth reflectance and shading for massive general natural scenes is challenging and even impossible. In this paper, we propose a nove…
The University of Texas at Austin（德克萨斯大学奥斯汀分校）；MIT-IBM Watson AI Lab（麻省理工学院-IBM沃森人工智能实验室）；Snap Inc.（Snap公司）
In this paper, we introduce a novel unsupervised domain adaptation technique for the task of 3D keypoint prediction from a single depth scan or image. Our key idea is to utilize the fact that predictions from different views of the same or similar objects should be consistent with each other. Such view consistency can provide effective regularization for keypoint prediction on unlabeled instances. In addition, we introduce a geometric alignment term to regularize predictions in the target domain. The resulting loss function can be effectively optimized via alternating minimization. We demonstr…
University of Tsukuba（筑波大学）；University of Tokyo（东京大学）；Gunosy Inc.（Gunosy公司）
Billions of live-streaming viewers share their opinions on scenes they are watching in real-time and interact with the event, commentators as well as other viewers via text comments. Thus, there is necessary to explore viewers' comments with scenes in E-sport live-streaming events. In this paper, we developed CS-lol, a new large-scale dataset containing comments from viewers paired with descriptions of game scenes in E-sports live-streaming. Moreover, we propose a task, namely viewer comment retrieval, to retrieve the viewer comments for the scene of the live-streaming event. Results on a seri…
The ability to decompose scenes in terms of abstract building blocks is crucial for general intelligence. Where those basic building blocks share meaningful properties, interactions and other regularities across scenes, such decompositions can simplify reasoning and facilitate imagination of novel scenarios. In particular, representing perceptual observations in terms of entities should improve data efficiency and transfer performance on a wide range of tasks. Thus we need models capable of discovering useful decompositions of scenes by identifying units with such regularities and representing…
Image synthesis driven by computer graphics achieved recently a remarkable realism, yet synthetic image data generated this way reveals a significant domain gap with respect to real-world data. This is especially true in autonomous driving scenarios, which represent a critical aspect for overcoming utilizing synthetic data for training neural networks. We propose a method based on domain-invariant scene representation to directly synthesize traffic scene imagery without rendering. Specifically, we rely on synthetic scene graphs as our internal representation and introduce an unsupervised neura…
In this work we present Discrete Attend Infer Repeat (Discrete-AIR), a Recurrent Auto-Encoder with structured latent distributions containing discrete categorical distributions, continuous attribute distributions, and factorised spatial attention. While inspired by the original AIR model andretaining AIR model's capability in identifying objects in an image, Discrete-AIR provides direct interpretability of the latent codes. We show that for Multi-MNIST and a multiple-objects version of dSprites dataset, the Discrete-AIR model needs just one categorical latent variable, one attribute variable (…
We present ObSuRF, a method which turns a single image of a scene into a 3D model represented as a set of Neural Radiance Fields (NeRFs), with each NeRF corresponding to a different object. A single forward pass of an encoder network outputs a set of latent vectors describing the objects in the scene. These vectors are used independently to condition a NeRF decoder, defining the geometry and appearance of each object. We make learning more computationally efficient by deriving a novel loss, which allows training NeRFs on RGB-D inputs without explicit ray marching. After confirming that the mod…
The ability to decompose complex multi-object scenes into meaningful abstractions like objects is fundamental to achieve higher-level cognition. Previous approaches for unsupervised object-oriented scene representation learning are either based on spatial-attention or scene-mixture approaches and limited in scalability which is a main obstacle towards modeling real-world scenes. In this paper, we propose a generative latent variable model, called SPACE, that provides a unified probabilistic modeling framework that combines the best of spatial-attention and scene-mixture approaches. SPACE can e…
ETH Zurich（苏黎世联邦理工学院）；University of Amsterdam（阿姆斯特丹大学）；KU Leuven（鲁汶大学）
Monocular depth reconstruction of complex and dynamic scenes is a highly challenging problem. While for rigid scenes learning-based methods have been offering promising results even in unsupervised cases, there exists little to no literature addressing the same for dynamic and deformable scenes. In this work, we present an unsupervised monocular framework for dense depth estimation of dynamic scenes, which jointly reconstructs rigid and non-rigid parts without explicitly modelling the camera motion. Using dense correspondences, we derive a training objective that aims to opportunistically pres…
Scene flow represents the motion of points in the 3D space, which is the counterpart of the optical flow that represents the motion of pixels in the 2D image. However, it is difficult to obtain the ground truth of scene flow in the real scenes, and recent studies are based on synthetic data for training. Therefore, how to train a scene flow network with unsupervised methods based on real-world data shows crucial significance. A novel unsupervised learning method for scene flow is proposed in this paper, which utilizes the images of two consecutive frames taken by monocular camera without the g…
We introduce the first approach to solve the challenging problem of unsupervised 4D visual scene understanding for complex dynamic scenes with multiple interacting people from multi-view video. Our approach simultaneously estimates a detailed model that includes a per-pixel semantically and temporally coherent reconstruction, together with instance-level segmentation exploiting photo-consistency, semantic and motion information. We further leverage recent advances in 3D pose estimation to constrain the joint semantic instance segmentation and 4D temporally coherent reconstruction. This enables…
University of Groningen（格罗宁根大学）；University of Barcelona（巴塞罗那大学）；Computer Vision Center（计算机视觉中心）
Nowadays, there is an upsurge of interest in using lifelogging devices. Such devices generate huge amounts of image data; consequently, the need for automatic methods for analyzing and summarizing these data is drastically increasing. We present a new method for familiar scene recognition in egocentric videos, based on background pattern detection through automatically configurable COSFIRE filters. We present some experiments over egocentric data acquired with the Narrative Clip.
Institute of Science Technology (IST) Austria（奥地利科学技术研究所）
Our goal in this work is to generate realistic videos given just one initial frame as input. Existing unsupervised approaches to this task do not consider the fact that a video typically shows a 3D environment, and that this should remain coherent from frame to frame even as the camera and objects move. We address this by developing a model that first estimates the latent 3D structure of the scene, including the segmentation of any moving objects. It then predicts future frames by simulating the object and camera dynamics, and rendering the resulting views. Importantly, it is trained end-to-en…
Shanghai Jiao Tong University（上海交通大学）；AI Institute, Shanghai Jiao Tong University（上海交通大学人工智能研究院）
Crowd segmentation is a fundamental task serving as the basis of crowded scene analysis, and it is highly desirable to obtain refined pixel-level segmentation maps. However, it remains a challenging problem, as existing approaches either require dense pixel-level annotations to train deep learning models or merely produce rough segmentation maps from optical or particle flows with physical models. In this paper, we propose the Motion Prior-Aware Siamese Network (MPASNET) for unsupervised crowd semantic segmentation. This model not only eliminates the need for annotation but also yields high-qu…
A natural approach to generative modeling of videos is to represent them as a composition of moving objects. Recent works model a set of 2D sprites over a slowly-varying background, but without considering the underlying 3D scene that gives rise to them. We instead propose to model a video as the view seen while moving through a scene with multiple 3D objects and a 3D background. Our model is trained from monocular videos without any supervision, yet learns to generate coherent 3D scenes containing several moving objects. We conduct detailed experiments on two datasets, going beyond the visual…
We tackle the problem of object-centric learning on point clouds, which is crucial for high-level relational reasoning and scalable machine intelligence. In particular, we introduce a framework, SPAIR3D, to factorize a 3D point cloud into a spatial mixture model where each component corresponds to one object. To model the spatial mixture model on point clouds, we derive the Chamfer Mixture Loss, which fits naturally into our variational training pipeline. Moreover, we adopt an object-specification scheme that describes each object's location relative to its local voxel grid cell. Such a scheme…
Deep learning has shown state-of-art classification performance on datasets such as ImageNet, which contain a single object in each image. However, multi-object classification is far more challenging. We present a unified framework which leverages the strengths of multiple machine learning methods, viz deep learning, probabilistic models and kernel methods to obtain state-of-art performance on Microsoft COCO, consisting of non-iconic images. We incorporate contextual information in natural images through a conditional latent tree probabilistic model (CLTM), where the object co-occurrences are …
Indian Institute of Engineering Science and Technology, Shibpur（印度希布尔工程科学技术学院）；Indian Institute of Space Science and Technology（印度空间科学技术学院）
We present an improved clustering based, unsupervised anomalous trajectory detection algorithm for crowded scenes. The proposed work is based on four major steps, namely, extraction of trajectories from crowded scene video, extraction of several features from these trajectories, independent mean-shift clustering and anomaly detection. First, the trajectories of all moving objects in a crowd are extracted using a multi feature video object tracker. These trajectories are then transformed into a set of feature spaces. Mean shift clustering is applied on these feature matrices to obtain distinct …
Institut national de la recherche scientifique (INRS-EMT)（国家科学研究院（INRS-EMT））
In this work, we introduce a two-step framework for generative modeling of temporal data. Specifically, the generative adversarial networks (GANs) setting is employed to generate synthetic scenes of moving objects. To do so, we propose a two-step training scheme within which: a generator of static frames is trained first. Afterwards, a recurrent model is trained with the goal of providing a sequence of inputs to the previously trained frames generator, thus yielding scenes which look natural. The adversarial setting is employed in both training steps. However, with the aim of avoiding known tr…
Shanghai Jiao Tong University（上海交通大学）；China University of Mining and Technology（中国矿业大学）
Scene flow represents the 3D motion of each point in the scene, which explicitly describes the distance and the direction of each point's movement. Scene flow estimation is used in various applications such as autonomous driving fields, activity recognition, and virtual reality fields. As it is challenging to annotate scene flow with ground truth for real-world data, this leaves no real-world dataset available to provide a large amount of data with ground truth for scene flow estimation. Therefore, many works use synthesized data to pre-train their network and real-world LiDAR data to finetune…
Australian National University（澳大利亚国立大学）；ACRV（澳大利亚机器人视觉中心）；Data61 CSIRO（联邦科学与工业研究组织数据61研究所）；NEC Labs America（NEC美国实验室）；Northwestern Polytechnical University（西北工业大学）
Unsupervised deep learning for optical flow computation has achieved promising results. Most existing deep-net based methods rely on image brightness consistency and local smoothness constraint to train the networks. Their performance degrades at regions where repetitive textures or occlusions occur. In this paper, we propose Deep Epipolar Flow, an unsupervised optical flow method which incorporates global geometric constraints into network learning. In particular, we investigate multiple ways of enforcing the epipolar constraint in flow estimation. To alleviate a "chicken-and-egg" type of pro…
To understand movies, humans constantly reason over the dialogues and actions shown in specific scenes and relate them to the overall storyline already seen. Inspired by this behaviour, we design ROLL, a model for knowledge-based video story question answering that leverages three crucial aspects of movie understanding: dialog comprehension, scene reasoning, and storyline recalling. In ROLL, each of these tasks is in charge of extracting rich and diverse information by 1) processing scene dialogues, 2) generating unsupervised video scene descriptions, and 3) obtaining external knowledge in a w…
Google Research（谷歌研究院）；Robotics at Google（谷歌机器人）；Waymo LLC（Waymo有限责任公司）
We present a method for jointly training the estimation of depth, ego-motion, and a dense 3D translation field of objects relative to the scene, with monocular photometric consistency being the sole source of supervision. We show that this apparently heavily underdetermined problem can be regularized by imposing the following prior knowledge about 3D translation fields: they are sparse, since most of the scene is static, and they tend to be constant for rigid moving objects. We show that this regularization alone is sufficient to train monocular depth prediction models that exceed the accuracy…
We study the problem of inferring an object-centric scene representation from a single image, aiming to derive a representation that explains the image formation process, captures the scene's 3D nature, and is learned without supervision. Most existing methods on scene decomposition lack one or more of these characteristics, due to the fundamental challenge in integrating the complex 3D-to-2D image formation process into powerful inference schemes like deep networks. In this paper, we propose unsupervised discovery of Object Radiance Fields (uORF), integrating recent progresses in neural 3D sc…
To help agents reason about scenes in terms of their building blocks, we wish to extract the compositional structure of any given scene (in particular, the configuration and characteristics of objects comprising the scene). This problem is especially difficult when scene structure needs to be inferred while also estimating the agent's location/viewpoint, as the two variables jointly give rise to the agent's observations. We present an unsupervised variational approach to this problem. Leveraging the shared structure that exists across different scenes, our model learns to infer two sets of lat…
Stanford University（斯坦福大学）；ETH Zürich（苏黎世联邦理工学院）；MPI for Intelligent Systems, Tübingen（图宾根马克斯·普朗克智能系统研究所）；KU Leuven（鲁汶大学）
Scene extrapolation -- the idea of generating novel views by flying into a given image -- is a promising, yet challenging task. For each predicted frame, a joint inpainting and 3D refinement problem has to be solved, which is ill posed and includes a high level of ambiguity. Moreover, training data for long-range scenes is difficult to obtain and usually lacks sufficient views to infer accurate camera poses. We introduce DiffDreamer, an unsupervised framework capable of synthesizing novel views depicting a long camera trajectory while training solely on internet-collected images of nature scen…
Nanyang Technological University（南洋理工大学）；The Chinese University of Hong Kong（香港中文大学）；A*STAR, Singapore（新加坡科技研究局）
Contrastive self-supervised learning has largely narrowed the gap to supervised pre-training on ImageNet. However, its success highly relies on the object-centric priors of ImageNet, i.e., different augmented views of the same image correspond to the same object. Such a heavily curated constraint becomes immediately infeasible when pre-trained on more complex scene images with many objects. To overcome this limitation, we introduce Object-level Representation Learning (ORL), a new self-supervised learning framework towards scene images. Our key insight is to leverage image-level self-supervise…
