Keyframe Extraction from Motion Capture Sequences with Graph based Deep Reinforcement Learning

Clinton Mo,Kun Hu,Shaohui Mei,Zebin Chen,Zhiyong Wang

DOI: https://doi.org/10.1145/3474085.3475635

2021-10-17

Abstract:Animation production workflows centred around motion capture techniques often require animators to edit the motion for various artistic and technical reasons. This process generally uses a set of keyframes. Unsupervised keyframe selection methods for motion capture sequences are highly demanded to reduce the laborious annotations. However, most existing methods are optimization-based, which cause the issues of flexibility and efficiency and eventually constrains the interactions and controls with animators. To address these limitations, we propose a novel graph based deep reinforcement learning method for efficient unsupervised keyframe selection. First, a reward function is devised in terms of reconstruction difference by comparing the original sequence and the interpolated sequence produced by the keyframes. The reward complies with the requirements of the animation pipeline satisfying: 1) incremental reward to evaluate the interpolated keyframes immediately; 2) order insensitivity for consistent evaluation; and 3) non-diminishing return for comparable rewards between optimal and sub-optimal solutions. Then by representing each skeleton frame as a graph, a graph-based deep agent is guided to heuristically select keyframes to maximize the reward. During the inference it is no longer necessary to estimate the reconstruction difference, and the evaluation time can be reduced significantly. The experimental results on the CMU Mocap dataset demonstrate that our proposed method is able to select keyframes at a high efficiency without clearly compromising the quality in comparison with the state-of-the-art methods.