4D-Animal: Freely Reconstructing Animatable 3D Animals from Videos

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zhong, Shanshan, Peng, Jiawei, Zheng, Zehan, Huang, Zhongzhan, Ma, Wufei, Zhang, Guofeng, Liu, Qihao, Yuille, Alan, Chen, Jieneng
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908449108918272
author Zhong, Shanshan
Peng, Jiawei
Zheng, Zehan
Huang, Zhongzhan
Ma, Wufei
Zhang, Guofeng
Liu, Qihao
Yuille, Alan
Chen, Jieneng
author_facet Zhong, Shanshan
Peng, Jiawei
Zheng, Zehan
Huang, Zhongzhan
Ma, Wufei
Zhang, Guofeng
Liu, Qihao
Yuille, Alan
Chen, Jieneng
contents Existing methods for reconstructing animatable 3D animals from videos typically rely on sparse semantic keypoints to fit parametric models. However, obtaining such keypoints is labor-intensive, and keypoint detectors trained on limited animal data are often unreliable. To address this, we propose 4D-Animal, a novel framework that reconstructs animatable 3D animals from videos without requiring sparse keypoint annotations. Our approach introduces a dense feature network that maps 2D representations to SMAL parameters, enhancing both the efficiency and stability of the fitting process. Furthermore, we develop a hierarchical alignment strategy that integrates silhouette, part-level, pixel-level, and temporal cues from pre-trained 2D visual models to produce accurate and temporally coherent reconstructions across frames. Extensive experiments demonstrate that 4D-Animal outperforms both model-based and model-free baselines. Moreover, the high-quality 3D assets generated by our method can benefit other 3D tasks, underscoring its potential for large-scale applications. The code is released at https://github.com/zhongshsh/4D-Animal.
format Preprint
id arxiv_https___arxiv_org_abs_2507_10437
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle 4D-Animal: Freely Reconstructing Animatable 3D Animals from Videos
Zhong, Shanshan
Peng, Jiawei
Zheng, Zehan
Huang, Zhongzhan
Ma, Wufei
Zhang, Guofeng
Liu, Qihao
Yuille, Alan
Chen, Jieneng
Computer Vision and Pattern Recognition
Existing methods for reconstructing animatable 3D animals from videos typically rely on sparse semantic keypoints to fit parametric models. However, obtaining such keypoints is labor-intensive, and keypoint detectors trained on limited animal data are often unreliable. To address this, we propose 4D-Animal, a novel framework that reconstructs animatable 3D animals from videos without requiring sparse keypoint annotations. Our approach introduces a dense feature network that maps 2D representations to SMAL parameters, enhancing both the efficiency and stability of the fitting process. Furthermore, we develop a hierarchical alignment strategy that integrates silhouette, part-level, pixel-level, and temporal cues from pre-trained 2D visual models to produce accurate and temporally coherent reconstructions across frames. Extensive experiments demonstrate that 4D-Animal outperforms both model-based and model-free baselines. Moreover, the high-quality 3D assets generated by our method can benefit other 3D tasks, underscoring its potential for large-scale applications. The code is released at https://github.com/zhongshsh/4D-Animal.
title 4D-Animal: Freely Reconstructing Animatable 3D Animals from Videos
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2507.10437