Book chapter
A Constrained Hybrid Optimization Algorithm for Morphable Appearance Models
Energy Minimization Methods in Computer Vision and Pattern Recognition, pp 568-583
2005
Abstract
In this paper, we propose a constrained hybrid optimization algorithm that incorporates several shape constraints into a gradient descent procedure using a novel unbiased cost function. Shape constraints are heuristically derived from face images where the face shape can be directly estimated based on ”motion” analysis. To better locate face contour points regardless of the background, local projection models are used. Experiments show that our algorithm benefits significantly from these shape constraints and achieves a much higher convergent rate compared to the inverse compositional optimization algorithm. We test our algorithm on different face databases, and demonstrate its robustness in presence of various illuminations, background patterns, as well as variations in face expressions.
Metrics
15 Record Views
Details
- Title
- A Constrained Hybrid Optimization Algorithm for Morphable Appearance Models
- Creators
- Cuiping Zhang - Drexel UniversityFernand S. Cohen - Drexel University
- Publication Details
- Energy Minimization Methods in Computer Vision and Pattern Recognition, pp 568-583
- Series
- Lecture Notes in Computer Science
- Publisher
- Springer Berlin Heidelberg; Berlin, Heidelberg
- Resource Type
- Book chapter
- Language
- English
- Academic Unit
- Electrical and Computer Engineering
- Web of Science ID
- WOS:000234193000037
- Scopus ID
- 2-s2.0-33646573975
- Other Identifier
- 991019170326304721
InCites Highlights
Data related to this publication, from InCites Benchmarking & Analytics tool:
- Web of Science research areas
- Computer Science, Artificial Intelligence
- Computer Science, Theory & Methods