Inferring 3D structure with a statistical image-based shape model
- 1 January 2003
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 641-647 vol.1
- https://doi.org/10.1109/iccv.2003.1238408
Abstract
We present an image-based approach to infer 3D structure parameters using a probabilistic "shape+structure" model. The 3D shape of an object class is represented by sets of contours from silhouette views simultaneously observed from multiple calibrated cameras, while structural features of interest on the object are denoted by a number of 3D locations. A prior density over the multiview shape and corresponding structure is constructed with a mixture of probabilistic principal components analyzers. Given a novel set of contours, we infer the unknown structure parameters from the new shape's Bayesian reconstruction. Model matching and parameter inference are done entirely in the image domain and require no explicit 3D construction. Our shape model enables accurate estimation of structure despite segmentation errors or missing views in the input silhouettes, and it works even with only a single input view. Using a training set of thousands of pedestrian images generated from a synthetic model, we can accurately infer the 3D locations of 19 joints on the body based on observed silhouette contours from real images.Keywords
This publication has 14 references indexed in Scilit:
- Eigen-points: control-point location using principal component analysesPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- Face recognition using eigenfacesPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- Multidimensional morphable modelsPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- Principal manifolds and probabilistic subspaces for visual recognitionPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- The dynamics of linear combinations: tracking 3D skeletons of human subjectsImage and Vision Computing, 2002
- Boundary finding with prior shape and smoothness modelsPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2000
- Image-based visual hullsPublished by Association for Computing Machinery (ACM) ,2000
- Mixtures of Probabilistic Principal Component AnalyzersNeural Computation, 1999
- Active Shape Models-Their Training and ApplicationComputer Vision and Image Understanding, 1995
- The visual hull concept for silhouette-based image understandingPublished by Institute of Electrical and Electronics Engineers (IEEE) ,1994