Automatically building appearance models from image sequences using salient features

Image and Vision Computing

Volumn 20, Issue 5-6, 2002, Pages 435-440

  Walker, K N ^a   Cootes, T F ^b   Taylor, C J ^b  

^a Image Metrics plc   (United Kingdom)

^b UNIVERSITY OF MANCHESTER   (United Kingdom)

Author keywords

Appearance model; Automatic correspondence; Automatic model building

Indexed keywords

FEATURE EXTRACTION; OBJECT RECOGNITION; PROBABILITY; STATISTICAL METHODS;

IMAGE SEQUENCES;

IMAGE UNDERSTANDING;

EID: 0037028833     PISSN: 02628856     EISSN: None     Source Type: Journal    
DOI: 10.1016/S0262-8856(02)00014-8     Document Type: Article

References (22)

1. Active shape models - Their training and application
2. Active appearance models
3. Automatic interpretation and coding of face images using flexible models
4. Correspondence based on distinct points using image invariants
5. Locating salient facial features using image invariants
6. Recognising and locating partially visible objects: The local-feature-focus method
7. Recognising partially occluded parts
8. A minimum description length approach to statistical shape modelling
9. Automatic landmark identification using a new method of non-rigid correspondence
10. Learning flexible models from image sequences
11. An adaptive eigenshape model
12. Coding facial expressions with Gabor wavelets
13. Tracking facial features using probabilistic network
14. Automatic learning of appearance face models
15. Shape and motion from image streams under orthography: A factorization method
16. Multidimensional morphable models: A framework for representing and matching object classes
17. A bootstrapping algorithm for learning linear models of object classes
18. Procrustes methods in the statistical analysis of shape
19. Eigenfaces for recognition
20. Locating salientobject features
21. Maintaining multiple motion model hypotheses over many views to recover matching and structure
22. Eigentracking: Robust matching and tracking of articulated objects using a view-based representation