|
Object Extraction Based On Active Contour |
INVESTIGATORS
Dr. Shen Dinggang, Dr. Horace H S Ip
BRIEF DESCRIPTION
In this project, we showed that existing shaped-based active contour models are not affine-invariant and we addressed the problem by presenting an affine-invariant snake model (AI-snake) such that its energy function are defined in terms local and global affine-invariant features. The main characteristic of the AI-snake is that, during the process of object extraction, the pose of the model contour is dynamically adjusted such that it is in alignment with the current snake contour by solving the snake-prototype correspondence problem and determining the required affine transformation. In addition, we formulate the correspondence matching between the snake and the object prototype as an error minimization process between two feature vectors which capture both local and global deformations information. We show that the technique is robust against object deformation and complex backgrounds.
FUNDING AGENCY
City University of Hong Kong
PUBLICATIONS
 | Horace H S Ip and Dinggong Shen, "An Affine Invariant Active Contour Model (AI-Snake) for Model-Based Segmentation", Image and Vision Computing, Vol. 16, pp. 135-146, 1998. |
| Dinggang Shen, Wai Him Wong and Horace H S Ip "Affine Invariant Images Retrieval by Correspondence Matching of Shapes", Image And Vision Computing, Vol. 17, No.7, pp. 489-499,1999. |
| "An affine-invariant active contour model (AI-snake) for model-based segmentation", Image and Vision Computing, 1997 (publication subject to some revisions). [Horace H S Ip and Dinggang Shen] |
| "Complex scene object tracking using using an affine invariant snake" (invited paper), Workshop on 3D Computer Vision'97, Hong Kong, May 17 1997. [Horace H S Ip and Dinggang Shen] |
| "Affine invariant shape matching for content-based retrieval", ICARCV'96, Singapore, Dec., 1996. [W H Wong, Dinggang Shen and Horace H S Ip] |
FIGURES OR IMAGES
