Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2511.13488 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912715173265408 |
|---|---|
| author | Wang, Lipeng Fan, Hongxing Chen, Haohua Huang, Zehuan Sheng, Lu |
| author_facet | Wang, Lipeng Fan, Hongxing Chen, Haohua Huang, Zehuan Sheng, Lu |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2511_13488 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | InterMoE: Individual-Specific 3D Human Interaction Generation via Dynamic Temporal-Selective MoE Wang, Lipeng Fan, Hongxing Chen, Haohua Huang, Zehuan Sheng, Lu Computer Vision and Pattern Recognition I.2.1 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. |
| title | InterMoE: Individual-Specific 3D Human Interaction Generation via Dynamic Temporal-Selective MoE |
| topic | Computer Vision and Pattern Recognition I.2.1 |
| url | https://arxiv.org/abs/2511.13488 |