Fast-SNARF: A Fast Deformer for Articulated Neural Fields
- 28 April 2023
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Pattern Analysis and Machine Intelligence
- Vol. 45 (10) , 11796-11809
Abstract
Neural fields have revolutionized the area of 3D reconstruction and novel view synthesis of rigid scenes. A key challenge in making such methods applicable to articulated objects, such as the human body, is to model the deformation of 3D locations between the rest pose (a canonical space) and the deformed space. We propose a new articulation module for neural fields, Fast-SNARF, which finds accurate correspondences between canonical space and posed space via iterative root finding. Fast-SNARF is a drop-in replacement in functionality to our previous work, SNARF, while significantly improving its computational efficiency. We contribute several algorithmic and implementation improvements over SNARF, yielding a speed-up of $150\times$ . These improvements include voxel-based correspondence search, pre-computing the linear blend skinning function, and an efficient software implementation with CUDA kernels. Fast-SNARF enables efficient and simultaneous optimization of shape and skinning weights given deformed observations without correspondences (e.g. 3D meshes). Because learning of deformation maps is a crucial component in many 3D human avatar methods and since Fast-SNARF provides a computationally efficient solution, we believe that this work represents a significant step towards the practical creation of 3D virtual humans.Keywords
All Related Versions
Funding Information
- Max Planck ETH Center for Learning Systems
- DFG EXC (390727645)
This publication has 59 references indexed in Scilit:
- Occupancy Networks: Learning 3D Reconstruction in Function Space2019
- Video Based Reconstruction of 3D People Models2018
- Dynamic FAUST: Registering Human Bodies in Motion2017
- 3D Menagerie: Modeling the 3D Shape and Pose of Animals2017
- SMPLACM Transactions on Graphics, 2015
- Pose-conditioned joint angle limits for 3D human pose reconstruction2015
- MoShACM Transactions on Graphics, 2014
- Skinning mesh animationsACM Transactions on Graphics, 2005
- A constructive geometry for computer graphicsThe Computer Journal, 1973
- A class of methods for solving nonlinear simultaneous equationsMathematics of Computation, 1965