Spatio-Temporal Analysis of Human Faces Using Multi-resolution Subdivision Surfaces
Proceedings of the 23rd DAGM-Symposium on Pattern Recognition
PCA based 3D shape reconstruction of human foot using multiple viewpoint cameras
ICVS'08 Proceedings of the 6th international conference on Computer vision systems
Hi-index | 0.00 |
This contribution describes a camera-based approach automatically create 3D models of persons. A setup of sixteen digital cameras is used to capture the images of the person to be scanned. Using a monochromatic background and the shape-from-silhouette approach a 3D model is created automatically. Using the original images, the models are realistically textured. In future 3D tele presence applications or multi-player games these models have to be animated using an internal skeleton structure. A principal component analysis is utilized to extract the skeleton automatically from the models. Compared to other full body scanning approaches no mechanically moving components are used, which makes the system especially suitable for low cost end user applications.