Unsupervised Shape and Pose Disentanglement for 3D Meshes
arXiv:2007.11341
Abstract
Parametric models of humans, faces, hands and animals have been widely used for a range of tasks such as image-based reconstruction, shape correspondence estimation, and animation. Their key strength is the ability to factor surface variations into shape and pose dependent components. Learning such models requires lots of expert knowledge and hand-defined object-specific constraints, making the learning approach unscalable to novel objects. In this paper, we present a simple yet effective approach to learn disentangled shape and pose representations in an unsupervised setting. We use a combination of self-consistency and cross-consistency constraints to learn pose and shape space from registered meshes. We additionally incorporate as-rigid-as-possible deformation(ARAP) into the training loop to avoid degenerate solutions. We demonstrate the usefulness of learned representations through a number of tasks including pose transfer and shape retrieval. The experiments on datasets of 3D humans, faces, hands and animals demonstrate the generality of our approach. Code is made available at https://virtualhumans.mpi-inf.mpg.de/unsup_shape_pose/.
References in corpus (8)
- Embodied Hands: Modeling and Capturing Hands and Bodies Together
- Disentangling factors of variation in deep representations using adversarial training
- A Unified Feature Disentangler for Multi-Domain Image Translation and Manipulation
- Deforming Autoencoders: Unsupervised Disentangling of Shape and Appearance
- Disentangled Human Body Embedding Based on Deep Hierarchical Neural Network
- Neural 3D Morphable Models: Spiral Convolutional Networks for 3D Shape Representation Learning and Generation
- Challenges in Disentangling Independent Factors of Variation
- Latent feature disentanglement for 3D meshes