Multi-view 3D human pose estimation combining single-frame recovery, temporal integration and model adaptation

Multi-view 3D human pose estimation combining single-frame recovery, temporal integration and model adaptation

Hofmann, K.M.
Gavrila, D.M.
TNO Defensie en Veiligheid

2009

We present a system for the estimation of unconstrained 3D human upper body movement from multiple cameras. Its main novelty lies in the integration of three components: single-frame pose recovery, temporal integration and model adaptation. Single-frame pose recovery consists of a hypothesis generation stage, where candidate 3D poses are generated based on hierarchical shape matching in the individual camera views. In the subsequent hypothesis verification stage, candidate 3D poses are re-projected to the other camera views and ranked according to a multi-view matching score. Temporal integration consists of computing best trajectories combining a motion model and observations in a Viterbi-style maximum likelihood approach. Poses that lie on the best trajectories are used to generate and adapt a texture model, which in turn enriches the shape component used for pose recovery. We demonstrate that our approach outperforms the state-of-the-art in experiments with large and challenging real-world data from an outdoor setting. The new data set is made public to facilitate benchmarking.

Image processing
Human pose estimation
Computer vision
3D mapping
Multiple cameras

http://resolver.tudelft.nl/uuid:8add2a0b-d77e-40e5-bdd1-27f5fa72bab1

https://doi.org/10.1109/cvprw.2009.5206508

248198

9781424439935

2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, CVPR Workshops 2009, 20 June - 25 June 2009, Miami, FL. USA, 2214-2221

5206508

conference paper