We present a deep, bidirectional, recurrent framework for cleaning noisy and incomplete motion capture data. It exploits temporal coherence and joint correlations to infer adaptive filters for each joint in each frame. A single model can be trained to denoise a heterogeneous mix of action types, under substantial amounts of noise. A signal that has both noise and gaps is preprocessed with a second bidirectional network that synthesizes missing frames from surrounding context. The approach handles a wide variety of noise types and long gaps, does not rely on knowledge of the noise distribution, and operates in a streaming setting. We validate our approach through extensive evaluations on noise both in joint angles and in joint positions, and show that it improves upon various alternatives.
@article{arxiv.1712.03380,
title = {A Deep Recurrent Framework for Cleaning Motion Capture Data},
author = {Utkarsh Mall and G. Roshan Lal and Siddhartha Chaudhuri and Parag Chaudhuri},
journal= {arXiv preprint arXiv:1712.03380},
year = {2017}
}