WildGHand: Learning Anti-Perturbation Gaussian Hand Avatars from Monocular In-the-Wild Videos

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Li, Hanhui, Huang, Xuan, Liu, Wanquan, Cheng, Yuhao, Chen, Long, Yan, Yiqiang, Liang, Xiaodan, Gao, Chenqiang
Natura: Preprint
Pubblicazione: 2026
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866910031307341824
author Li, Hanhui
Huang, Xuan
Liu, Wanquan
Cheng, Yuhao
Chen, Long
Yan, Yiqiang
Liang, Xiaodan
Gao, Chenqiang
author_facet Li, Hanhui
Huang, Xuan
Liu, Wanquan
Cheng, Yuhao
Chen, Long
Yan, Yiqiang
Liang, Xiaodan
Gao, Chenqiang
contents Despite recent progress in 3D hand reconstruction from monocular videos, most existing methods rely on data captured in well-controlled environments and therefore degrade in real-world settings with severe perturbations, such as hand-object interactions, extreme poses, illumination changes, and motion blur. To tackle these issues, we introduce WildGHand, an optimization-based framework that enables self-adaptive 3D Gaussian splatting on in-the-wild videos and produces high-fidelity hand avatars. WildGHand incorporates two key components: (i) a dynamic perturbation disentanglement module that explicitly represents perturbations as time-varying biases on 3D Gaussian attributes during optimization, and (ii) a perturbation-aware optimization strategy that generates per-frame anisotropic weighted masks to guide optimization. Together, these components allow the framework to identify and suppress perturbations across both spatial and temporal dimensions. We further curate a dataset of monocular hand videos captured under diverse perturbations to benchmark in-the-wild hand avatar reconstruction. Extensive experiments on this dataset and two public datasets demonstrate that WildGHand achieves state-of-the-art performance and substantially improves over its base model across multiple metrics (e.g., up to a $15.8\%$ relative gain in PSNR and a $23.1\%$ relative reduction in LPIPS). Our implementation and dataset are available at https://github.com/XuanHuang0/WildGHand.
format Preprint
id arxiv_https___arxiv_org_abs_2602_20556
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle WildGHand: Learning Anti-Perturbation Gaussian Hand Avatars from Monocular In-the-Wild Videos
Li, Hanhui
Huang, Xuan
Liu, Wanquan
Cheng, Yuhao
Chen, Long
Yan, Yiqiang
Liang, Xiaodan
Gao, Chenqiang
Computer Vision and Pattern Recognition
Despite recent progress in 3D hand reconstruction from monocular videos, most existing methods rely on data captured in well-controlled environments and therefore degrade in real-world settings with severe perturbations, such as hand-object interactions, extreme poses, illumination changes, and motion blur. To tackle these issues, we introduce WildGHand, an optimization-based framework that enables self-adaptive 3D Gaussian splatting on in-the-wild videos and produces high-fidelity hand avatars. WildGHand incorporates two key components: (i) a dynamic perturbation disentanglement module that explicitly represents perturbations as time-varying biases on 3D Gaussian attributes during optimization, and (ii) a perturbation-aware optimization strategy that generates per-frame anisotropic weighted masks to guide optimization. Together, these components allow the framework to identify and suppress perturbations across both spatial and temporal dimensions. We further curate a dataset of monocular hand videos captured under diverse perturbations to benchmark in-the-wild hand avatar reconstruction. Extensive experiments on this dataset and two public datasets demonstrate that WildGHand achieves state-of-the-art performance and substantially improves over its base model across multiple metrics (e.g., up to a $15.8\%$ relative gain in PSNR and a $23.1\%$ relative reduction in LPIPS). Our implementation and dataset are available at https://github.com/XuanHuang0/WildGHand.
title WildGHand: Learning Anti-Perturbation Gaussian Hand Avatars from Monocular In-the-Wild Videos
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2602.20556