Towards 3D-Aware Video Diffusion Models: Render-Free Human Motion Control with Mesh Tokenization
2026-06-01 • Computer Vision and Pattern Recognition
Computer Vision and Pattern RecognitionArtificial Intelligence
AI summaryⓘ
The authors explore whether video diffusion models truly understand the 3D shape and movement of humans instead of just creating realistic 2D videos. They introduce a new method that uses compressed 3D human body data (mesh tokens) directly to guide video creation, rather than using 2D motion videos. This approach helps the model learn about the 3D structure, camera angles, and movements together. Their tests show improved results and fewer mistakes compared to older methods. This suggests these models can better grasp complex 3D human motions with their new technique.
diffusion models3D human meshvideo generationmotion controltokenizationcamera viewpointDiT architecture3D geometrypose editingmotion tokens
Authors
Jingyun Liang, Min Wei, Shikai Li, Yizeng Han, Hangjie Yuan, Lei Sun, Weihua Chen, Fan Wang
Abstract
Diffusion models have shown remarkable success in video generation. However, whether such models are truly aware of the 3D structure underlying visual observations, rather than simply reproducing plausible 2D projections, remains an open question. In this work, we investigate this question through human motion control, a task that requires precise modelling of 3D human geometry, motion, camera viewpoint, and scene context. Unlike prior methods that rely on rendered 2D motion guidance videos, we propose a render-free framework that conditions video generation directly on compressed 3D human mesh tokens. This representation preserves full 3D geometric information while enabling a unified token-based generation pipeline that processes video tokens jointly with motion tokens in a DiT-based architecture. This design requires the model to reason jointly about appearance, 3D structure, and camera viewpoint during video generation. Experimental results demonstrate strong performance on human motion control benchmarks, while reducing artifacts induced by view-dependent 2D guidance and trajectory-pose mismatches during editing. These findings suggest that video diffusion models, when equipped with mesh tokenization, can better capture complex 3D human structures and their interactions with the surrounding environment.