Unsupervised Learning of Probably Symmetric Deformable 3D Objects from Images in the Wild
arXiv:1911.11130
Abstract
We propose a method to learn 3D deformable object categories from raw single-view images, without external supervision. The method is based on an autoencoder that factors each input image into depth, albedo, viewpoint and illumination. In order to disentangle these components without supervision, we use the fact that many object categories have, at least in principle, a symmetric structure. We show that reasoning about illumination allows us to exploit the underlying object symmetry even if the appearance is not symmetric due to shading. Furthermore, we model objects that are probably, but not certainly, symmetric by predicting a symmetry probability map, learned end-to-end with the other components of the model. Our experiments show that this method can recover very accurately the 3D shape of human faces, cat faces and cars from single-view images, without any supervision or a prior shape model. On benchmarks, we demonstrate superior accuracy compared to another method that uses supervision at the level of 2D image correspondences.
arXiv commentCVPR 2020 Oral. Project page: https://elliottwu.com/projects/unsup3d/
References in corpus (11)
- What Uncertainties Do We Need in Bayesian Deep Learning for Computer Vision?
- VoxCeleb2: Deep Speaker Recognition
- Depth Map Prediction from a Single Image using a Multi-Scale Deep Network
- Learning a Probabilistic Latent Space of Object Shapes via 3D Generative-Adversarial Modeling
- DeMoN: Depth and Motion Network for Learning Monocular Stereo
- Learning Representations and Generative Models for 3D Point Clouds
- Learning to Predict 3D Objects with an Interpolation-based Differentiable Renderer
- Discovery of Latent 3D Keypoints via End-to-end Geometric Reasoning
- Soft Rasterizer: A Differentiable Renderer for Image-based 3D Reasoning
- Deforming Autoencoders: Unsupervised Disentangling of Shape and Appearance
- Unsupervised Generative 3D Shape Learning from Natural Images