Accurate 3D human pose estimation from single images is possible with sophisticated deep-net architectures that have been trained on very large datasets. However, this still leaves open the problem of capturing motions for which no such database exists. Manual annotation is tedious, slow, and error-prone. In this paper, we propose to replace most of the annotations by the use of multiple views, at training time only. Specifically, we train the system to predict the same pose in all views. Such a consistency constraint is necessary but not sufficient to predict accurate poses. We therefore complement it with a supervised loss aiming to predict the correct pose in a small set of labeled images, and with a regularization term that penalizes drift from initial predictions. Furthermore, we propose a method to estimate camera pose jointly with human pose, which lets us utilize multiview footage where calibration is difficult, e.g., for pan-tilt or moving handheld cameras. We demonstrate the effectiveness of our approach on established benchmarks, as well as on a new Ski dataset with rotating cameras and expert ski motion, for which annotations are truly hard to obtain.would be to annotate video data. However, achieving high accuracy would require a great deal of annotation, which is tedious, slow, and error-prone. As illustrated by Fig. 1, we therefore propose to replace most of the annotations by the use of multiple views, at training time only. Specifically, we use them to provide weak supervision and force the system to predict the same pose in all views.While such view consistency constraints increase accuracy, they are unfortunately not sufficient. For example, the network can learn to always predict the same pose, independently of the input image. To prevent this, we use a small set of images with ground-truth poses, which serve a dual purpose. First, they provide strong supervision during training. Second, they let us regularize the multi-view predictions by encouraging them to remain close to the predictions of a network trained with the scarce supervised data only.In addition, we propose to use a normalized pose distance to evaluate all losses involving poses. It disentangles pose from scale, and we found it to be key to maintain accuracy when the annotated data is scarce.Our experiments demonstrate the effectiveness of our weakly-supervised multi-view training strategy on several 1 arXiv:1803.04775v2 [cs.CV]

While Physics-Based Simulation (PBS) can accurately drape a 3D garment on a 3D body, it remains too costly for real-time applications, such as virtual try-on. By contrast, inference in a deep network, requiring a single forward pass, is much faster. Taking advantage of this, we propose a novel architecture to fit a 3D garment template to a 3D body. Specifically, we build upon the recent progress in 3D point cloud processing with deep networks to extract garment features at varying levels of detail, including pointwise, patch-wise and global features. We fuse these features with those extracted in parallel from the 3D body, so as to model the cloth-body interactions. The resulting two-stream architecture, which we call as GarNet, is trained using a loss function inspired by physics-based modeling, and delivers visually plausible garment shapes whose 3D points are, on average, less than 1 cm away from those of a PBS method, while running 100 times faster. Moreover, the proposed method can model various garment types with different cutting patterns when parameters of those patterns are given as input to the network.

Learning general image representations has proven key to the success of many computer vision tasks. For example, many approaches to image understanding problems rely on deep networks that were initially trained on ImageNet, mostly because the learned features are a valuable starting point to learn from limited labeled data. However, when it comes to 3D motion capture of multiple people, these features are only of limited use.In this paper, we therefore propose an approach to learning features that are useful for this purpose. To this end, we introduce a self-supervised approach to learning what we call a neural scene decomposition (NSD) that can be exploited for 3D pose estimation. NSD comprises three layers of abstraction to represent human subjects: spatial layout in terms of bounding-boxes and relative depth; a 2D shape representation in terms of an instance segmentation mask; and subject-specific appearance and 3D pose information. By exploiting self-supervision coming from multiview data, our NSD model can be trained end-to-end without any 2D or 3D supervision. In contrast to previous approaches, it works for multiple persons and full-frame images. Because it encodes 3D geometry, NSD can then be effectively leveraged to train a 3D pose estimation network from small amounts of annotated data.

No abstract

While supervised object detection and segmentation methods achieve impressive accuracy, they generalize poorly to images whose appearance significantly differs from the data they have been trained on. To address this when annotating data is prohibitively expensive, we introduce a self-supervised detection and segmentation approach that can work with single images captured by a potentially moving camera. At the heart of our approach lies the observation that object segmentation and background reconstruction are linked tasks, and that, for structured scenes, background regions can be re-synthesized from their surroundings, whereas regions depicting the moving object cannot. We encode this intuition into a self-supervised loss function that we exploit to train a proposal-based segmentation network. To account for the discrete nature of the proposals, we develop a Monte Carlo-based training strategy that allows the algorithm to explore the large space of object proposals. We apply our method to human detection and segmentation in images that visually depart from those of standard benchmarks and outperform existing self-supervised methods.