Representing Animatable Avatar via Factorized Neural Fields
Open Access
- 28 August 2025
- journal article
- Published by Wiley in Computer Graphics Forum
- Vol. 44 (5)
Abstract
For reconstructing high‐fidelity human 3D models from monocular videos, it is crucial to maintain consistent large‐scale body shapes along with finely matched subtle wrinkles. This paper explores how per‐frame rendering results can be factorized into a pose‐independent component and a corresponding pose‐dependent counterpart to facilitate frame consistency at multiple scales. Pose adaptive texture features are further improved by restricting the frequency bands of these two components. Pose‐independent outputs are expected to be low‐frequency, while high‐frequency information is linked to pose‐dependent factors. We implement this with a dual‐branch network. The first branch takes coordinates in the canonical space as input, while the second one additionally considers features outputted by the first branch and pose information of each frame. A final network integrates the information predicted by both branches and utilizes volume rendering to generate photo‐realistic 3D human images. Through experiments, we demonstrate that our method consistently surpasses all state‐of‐the‐art methods in preserving high‐frequency details and ensuring consistent body contours. Our code is accessible at https://github.com/ChunjinSong/facavatar.Keywords
All Related Versions
Funding Information
- Natural Sciences and Engineering Research Council of Canada
- Maryland Advanced Research Computing Center
- Vector Institute
- Canadian Institute for Advanced Research
- Natural Sciences and Engineering Research Council of Canada
- Knut och Alice Wallenbergs Stiftelse
This publication has 51 references indexed in Scilit:
- TensoRF: Tensorial Radiance Fields2022
- TAVA: Template-free Animatable Volumetric Actors2022
- NeuMan: Neural Human Radiance Field from a Single Video2022
- ARAH: Animatable Volume Rendering of Articulated Human SDFs2022
- NeRF: Representing Scenes as Neural Radiance Fields for View Synthesis2020
- DeepSDF: Learning Continuous Signed Distance Functions for Shape Representation2019
- Diverse Image-to-Image Translation via Disentangled Representations2018
- The Unreasonable Effectiveness of Deep Features as a Perceptual Metric2018
- General Automatic Human Shape and Motion Capture Using Volumetric Contour Cues2016
- Determining optical flowArtificial Intelligence, 1981