Human Motion Synthesis in 3D Scenes via Unified Scene Semantic Occupancy
Proceedings of the AAAI Conference on Artificial Intelligence · 14.03.2026 · 40 (6) · 4248-4256
DOI: 10.1609/aaai.v40i6.42421 ↗Kurz gefasst
Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 14.03.2026. Verfasst von Jingyu Gong und 7 weiteren Autor:innen.
Abstract (Originalsprache)
Human motion synthesis in 3D scenes relies heavily on scene comprehension, while current methods focus mainly on scene structure but ignore the semantic understanding. In this paper, we propose a human motion synthesis framework that take an unified Scene Semantic Occupancy (SSO) for scene representation, termed SSOMotion. We design a bi-directional tri-plane decomposition to derive a compact version of the SSO, and scene semantics are mapped to an unified feature space via CLIP encoding and shared linear dimensionality reduction. Such strategy can derive the fine-grained scene semantic structures while significantly reduce redundant computations. We further take these scene hints and movement direction derived from instructions for motion control via frame-wise scene query. Extensive experiments and ablation studies conducted on cluttered scenes using ShapeNet furniture, as well as scanned scenes from PROX and Replica datasets, demonstrate its cutting-edge performance while validating its effectiveness and generalization ability.
Zitieren
APA
Gong, J., Tong, K., Chen, Z., Yuan, C., Chen, M., Zhang, Z., Tan, X., & Xie, Y. (2026). Human Motion Synthesis in 3D Scenes via Unified Scene Semantic Occupancy. Proceedings of the AAAI Conference on Artificial Intelligence, 40(6), 4248-4256. https://doi.org/10.1609/aaai.v40i6.42421
Vancouver
Gong J, Tong K, Chen Z, Yuan C, Chen M, Zhang Z, et al. Human Motion Synthesis in 3D Scenes via Unified Scene Semantic Occupancy. Proceedings of the AAAI Conference on Artificial Intelligence. 2026;40(6):4248-4256. doi:10.1609/aaai.v40i6.42421
BibTeX
@article{Gong2026,
author = {Gong, Jingyu and Tong, Kunkun and Chen, Zhuoran and Yuan, Chuanhan and Chen, Mingang and Zhang, Zhizhong and Tan, Xin and Xie, Yuan},
title = {Human Motion Synthesis in 3D Scenes via Unified Scene Semantic Occupancy},
journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
year = {2026},
volume = {40},
number = {6},
pages = {4248--4256},
doi = {10.1609/aaai.v40i6.42421}
}
EUVIMED