A Deep Recurrent Framework for Cleaning Motion Capture Data
This addresses the problem of cleaning motion capture data for applications in animation, gaming, and biomechanics, but it is incremental as it builds on existing deep learning and recurrent network techniques for denoising.
The paper tackles the problem of cleaning noisy and incomplete motion capture data by developing a deep bidirectional recurrent framework that exploits temporal coherence and joint correlations to infer adaptive filters. The result is a single model that can denoise heterogeneous action types under substantial noise, handle various noise types and long gaps without relying on noise distribution knowledge, and operate in streaming settings, showing improvements over alternatives in evaluations on joint angles and positions.
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.