COIN:Control-Inpainting Diffusion Prior for Human and Camera Motion Estimation

COIN
Capturing global human and camera motion from a dynamic camera presents unique challenges. In the input video, a person is riding a skateboard – while the local body motion may remain relatively constant, the global position of the individual changes significantly. Current state-of-the-art methods such as PACE and WHAM fail catastrophically on such out-of-distribution motions. Our approach, COIN, gracefully handles such challenging cases, owing to our control-inpainting motion diffusion prior and novel human-scene relation loss.

Abstract

Estimating global human motion from moving cameras is challenging due to the entanglement of human and camera motions. To mitigate the ambiguity, existing methods leverage learned human motion priors, which however often result in oversmoothed motions with misaligned 2D projections. To tackle this problem, we propose COIN, a control-inpainting motion diffusion prior that enables fine-grained control to disentangle human and camera motions. Although pre-trained motion diffusion models encode rich motion priors, we find it non-trivial to leverage such knowledge to guide global motion estimation from RGB videos. COIN introduces a novel control-inpainting score distillation sampling method to ensure well-aligned, consistent, and high-quality motion from the diffusion prior within a joint optimization framework. Furthermore, we introduce a new human-scene relation loss to alleviate the scale ambiguity by enforcing consistency among the humans, camera, and scene. Experiments on three challenging benchmarks demonstrate the effectiveness of COIN, which outperforms the state-of-the-art methods in terms of global human motion estimation and camera motion estimation. As an illustrative example, COIN outperforms the state-of-the-art method by 33% in world joint position error (W-MPJPE) on the RICH dataset.

Overview

COIN Overview

Results

Citation 

            
@inproceedings{li2024coin,
    title={COIN: Control-Inpainting Diffusion Prior for Human and Camera Motion Estimation},
    author={Li, Jiefeng and Yuan, Ye, and Rempe, Davis and Zhang, Haotian and Lu, Cewu and Kautz, Jan and Iqbal, Umar},
    booktitle={European Conference on Computer Vision (ECCV)},
    year={2024}
}