Title: REWIND: Real-Time Egocentric Whole-Body Motion Diffusion with Exemplar-Based Identity Conditioning Show Chinese title

Title: 重放：基于实例的身份条件的自中心全身运动扩散的实时技术

Abstract: We present REWIND (Real-Time Egocentric Whole-Body Motion Diffusion), a one-step diffusion model for real-time, high-fidelity human motion estimation from egocentric image inputs. While an existing method for egocentric whole-body (i.e., body and hands) motion estimation is non-real-time and acausal due to diffusion-based iterative motion refinement to capture correlations between body and hand poses, REWIND operates in a fully causal and real-time manner. To enable real-time inference, we introduce (1) cascaded body-hand denoising diffusion, which effectively models the correlation between egocentric body and hand motions in a fast, feed-forward manner, and (2) diffusion distillation, which enables high-quality motion estimation with a single denoising step. Our denoising diffusion model is based on a modified Transformer architecture, designed to causally model output motions while enhancing generalizability to unseen motion lengths. Additionally, REWIND optionally supports identity-conditioned motion estimation when identity prior is available. To this end, we propose a novel identity conditioning method based on a small set of pose exemplars of the target identity, which further enhances motion estimation quality. Through extensive experiments, we demonstrate that REWIND significantly outperforms the existing baselines both with and without exemplar-based identity conditioning. Show Chinese abstract

Abstract: 我们提出了REWIND（自中心实时全身运动扩散），这是一种一步扩散模型，用于从自中心图像输入实时、高保真地估计人体运动。 现有的自中心全身（即身体和手部）运动估计方法由于基于扩散的迭代运动细化来捕捉身体和手部姿态之间的相关性，因此是非实时且非因果的。而REWIND以完全因果和实时的方式运行。 为了实现实时推理，我们引入了（1）级联的身体-手部去噪扩散，它以快速前馈的方式有效地建模了自中心身体和手部运动之间的相关性，以及（2）扩散蒸馏，它通过单一步骤的去噪实现了高质量的运动估计。 我们的去噪扩散模型基于修改后的Transformer架构，旨在因果建模输出运动的同时增强对未知运动长度的泛化能力。 此外，当身份先验可用时，REWIND可选地支持基于身份的运动估计。 为此，我们提出了一种基于目标身份的一小组姿态样本的新颖身份条件方法，这进一步提高了运动估计的质量。 通过广泛的实验，我们证明了REWIND无论是否基于样本的身份条件，在性能上都显著优于现有基线。