无需配对数据，用扩散模型生成自然且符合场景的人体动作。
Diffusion Implicit Policy for Unpaired Scene-aware Motion Synthesis
- 训练时解耦人体与场景交互，推理时引入隐式策略指导扩散过程
- 在ShapeNet、PROX和Replica数据集上动作自然度和交互合理性均优于现有方法
- 适合需要泛化到多种场景的运动生成任务，尤其适用于无配对数据场景
由于广泛应用，场景感知运动合成近年来受到广泛关注。现有方法高度依赖配对的动作-场景数据，当仅在少数特定场景上训练时难以泛化。为此，我们提出统一框架Diffusion Implicit Policy（DIP），不再需要配对数据。本文在训练中将人体-场景交互与运动合成解耦，并在推理阶段引入基于交互的隐式策略作用于运动扩散过程。通过迭代去噪与隐式策略优化，可同时保持动作自然性和交互合理性。针对长期运动生成，我们在联合旋转幂空间中引入动作融合机制。在ShapeNet家具场景及PROX和Replica真实场景上的实验表明，本框架在动作自然度和交互合理性方面均优于前沿方法，证明了DIP在更广泛任务和多样化场景中进行运动合成的可行性。代码将公开于https://github.com/jingyugong/DIP。
原文摘要 · Abstract (English)
Scene-aware motion synthesis has been widely researched recently due to its numerous applications. Prevailing methods rely heavily on paired motion-scene data, while it is difficult to generalize to diverse scenes when trained only on a few specific ones. Thus, we propose a unified framework, termed Diffusion Implicit Policy (DIP), for scene-aware motion synthesis, where paired motion-scene data are no longer necessary. In this paper, we disentangle human-scene interaction from motion synthesis during training, and then introduce an interaction-based implicit policy into motion diffusion during inference. Synthesized motion can be derived through iterative diffusion denoising and implicit policy optimization, thus motion naturalness and interaction plausibility can be maintained simultaneously. For long-term motion synthesis, we introduce motion blending in joint rotation power space. The proposed method is evaluated on synthesized scenes with ShapeNet furniture, and real scenes from PROX and Replica. Results show that our framework presents better motion naturalness and interaction plausibility than cutting-edge methods. This also indicates the feasibility of utilizing the DIP for motion synthesis in more general tasks and versatile scenes. Code will be publicly available at https://github.com/jingyugong/DIP.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。