InterMoE: Individual-Specific 3D Human Interaction Generation via Dynamic Temporal-Selective MoE
Proceedings of the AAAI Conference on Artificial Intelligence · 14.03.2026 · 40 (3) · 2137-2145
DOI: 10.1609/aaai.v40i3.37196 ↗Kurz gefasst
Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 14.03.2026. Verfasst von Lipeng Wang und 4 weiteren Autor:innen. Die Arbeit wurde bisher 3-mal zitiert (Crossref).
Abstract (Originalsprache)
Generating high-quality human interactions holds significant value for applications like virtual reality and robotics. However, existing methods often fail to preserve unique individual characteristics or fully adhere to textual descriptions. To address these challenges, we introduce InterMoE, a novel framework built on a Dynamic Temporal-Selective Mixture of Experts. The core of InterMoE is a routing mechanism that synergistically uses both high-level text semantics and low-level motion context to dispatch temporal motion features to specialized experts. This allows experts to dynamically determine the selection capacity and focus on critical temporal features, thereby preserving specific individual characteristic identities while ensuring high semantic fidelity. Extensive experiments show that InterMoE achieves state-of-the-art performance in individual-specific high-fidelity 3D human interaction generation, reducing FID scores by 9% on the InterHuman dataset and 22% on InterX.
Zitieren
APA
Wang, L., Fan, H., Chen, H., Huang, Z., & Sheng, L. (2026). InterMoE: Individual-Specific 3D Human Interaction Generation via Dynamic Temporal-Selective MoE. Proceedings of the AAAI Conference on Artificial Intelligence, 40(3), 2137-2145. https://doi.org/10.1609/aaai.v40i3.37196
Vancouver
Wang L, Fan H, Chen H, Huang Z, Sheng L. InterMoE: Individual-Specific 3D Human Interaction Generation via Dynamic Temporal-Selective MoE. Proceedings of the AAAI Conference on Artificial Intelligence. 2026;40(3):2137-2145. doi:10.1609/aaai.v40i3.37196
BibTeX
@article{Wang2026,
author = {Wang, Lipeng and Fan, Hongxing and Chen, Haohua and Huang, Zehuan and Sheng, Lu},
title = {InterMoE: Individual-Specific 3D Human Interaction Generation via Dynamic Temporal-Selective MoE},
journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
year = {2026},
volume = {40},
number = {3},
pages = {2137--2145},
doi = {10.1609/aaai.v40i3.37196}
}
EUVIMED