Saved in:
Bibliographic Details
Main Authors: Xia, Qi, Cong, Peishan, Wang, Ziyi, Sun, Yujing, Sun, Qin, Zhu, Xinge, Ye, Mao, Yang, Ruigang, Ma, Yuexin
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2604.13581
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908965232705536
author Xia, Qi
Cong, Peishan
Wang, Ziyi
Sun, Yujing
Sun, Qin
Zhu, Xinge
Ye, Mao
Yang, Ruigang
Ma, Yuexin
author_facet Xia, Qi
Cong, Peishan
Wang, Ziyi
Sun, Yujing
Sun, Qin
Zhu, Xinge
Ye, Mao
Yang, Ruigang
Ma, Yuexin
contents Accurately reconstructing human behavior in close-interaction scenarios is crucial for enabling realistic virtual interactions in augmented reality, precise motion analysis in sports, and natural collaborative behavior in human-robot tasks. Reliable reconstruction in these contexts significantly enhances the realism and effectiveness of AI-driven interactive applications. However, human reconstruction from monocular videos in close-interaction scenarios remains challenging due to severe mutual occlusions, leading local motion ambiguity, disrupted temporal continuity and spatial relationship error. In this paper, we propose SocialMirror, a diffusion-based framework that integrates semantic and geometric cues to effectively address these issues. Specifically, we first leverage high-level interaction descriptions generated by a vision-language model to guide a semantic-guided motion infiller, hallucinating occluded bodies and resolving local pose ambiguities. Next, we propose a sequence-level temporal refiner that enforces smooth, jitter-free motions, while incorporating geometric constraints during sampling to ensure plausible contact and spatial relationships. Evaluations on multiple interaction benchmarks show that SocialMirror achieves state-of-the-art performance in reconstructing interactive human meshes, demonstrating strong generalization across unseen datasets and in-the-wild scenarios. The code will be released upon publication.
format Preprint
id arxiv_https___arxiv_org_abs_2604_13581
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle SocialMirror: Reconstructing 3D Human Interaction Behaviors from Monocular Videos with Semantic and Geometric Guidance
Xia, Qi
Cong, Peishan
Wang, Ziyi
Sun, Yujing
Sun, Qin
Zhu, Xinge
Ye, Mao
Yang, Ruigang
Ma, Yuexin
Computer Vision and Pattern Recognition
Accurately reconstructing human behavior in close-interaction scenarios is crucial for enabling realistic virtual interactions in augmented reality, precise motion analysis in sports, and natural collaborative behavior in human-robot tasks. Reliable reconstruction in these contexts significantly enhances the realism and effectiveness of AI-driven interactive applications. However, human reconstruction from monocular videos in close-interaction scenarios remains challenging due to severe mutual occlusions, leading local motion ambiguity, disrupted temporal continuity and spatial relationship error. In this paper, we propose SocialMirror, a diffusion-based framework that integrates semantic and geometric cues to effectively address these issues. Specifically, we first leverage high-level interaction descriptions generated by a vision-language model to guide a semantic-guided motion infiller, hallucinating occluded bodies and resolving local pose ambiguities. Next, we propose a sequence-level temporal refiner that enforces smooth, jitter-free motions, while incorporating geometric constraints during sampling to ensure plausible contact and spatial relationships. Evaluations on multiple interaction benchmarks show that SocialMirror achieves state-of-the-art performance in reconstructing interactive human meshes, demonstrating strong generalization across unseen datasets and in-the-wild scenarios. The code will be released upon publication.
title SocialMirror: Reconstructing 3D Human Interaction Behaviors from Monocular Videos with Semantic and Geometric Guidance
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2604.13581