Abstract | ||
---|---|---|
We present a robust method for capturing articulated hand motions in realtime using a single depth camera. Our system is based on a realtime registration process that accurately reconstructs hand poses by fitting a 3D articulated hand model to depth images. We register the hand model using depth, silhouette, and temporal information. To effectively map low-quality depth maps to realistic hand poses, we regularize the registration with kinematic and temporal priors, as well as a data-driven prior built from a database of realistic hand poses. We present a principled way of integrating such priors into our registration optimization to enable robust tracking without severely restricting the freedom of motion. A core technical contribution is a new method for computing tracking correspondences that directly models occlusions typical of single-camera setups. To ensure reproducibility of our results and facilitate future research, we fully disclose the source code of our implementation. |
Year | DOI | Venue |
---|---|---|
2015 | 10.1111/cgf.12700 | COMPUTER GRAPHICS FORUM |
Field | DocType | Volume |
Computer vision,Motion capture,Kinematics,Computer graphics (images),Computer science,Source code,Silhouette,Artificial intelligence,Prior probability | Journal | 34.0 |
Issue | ISSN | Citations |
5.0 | 0167-7055 | 52 |
PageRank | References | Authors |
1.55 | 29 | 6 |
Name | Order | Citations | PageRank |
---|---|---|---|
Andrea Tagliasacchi | 1 | 716 | 31.90 |
Matthias Schröder | 2 | 74 | 4.09 |
Anastasia Tkach | 3 | 52 | 1.55 |
Sofien Bouaziz | 4 | 934 | 35.79 |
Mario Botsch | 5 | 2385 | 116.10 |
Mark Pauly | 6 | 4970 | 201.49 |