DocumentCode :
1944349
Title :
A Model-Based Interactive Object Segmentation Procedure
Author :
Zou, Jie
Author_Institution :
Dept. of Electr. Comput. & Sci. Eng., Rensselaer Polytech. Inst., Troy, NY
Volume :
1
fYear :
2005
fDate :
5-7 Jan. 2005
Firstpage :
522
Lastpage :
527
Abstract :
A global parametric shape model (boundary) of the object is optimized according to evidence accumulated from local features and the prior probability of the model parameters learned from already segmented training samples. The parametric boundary is then deformed to the most conspicuous edge pixels near it. We emphasize the effectiveness of human intervention. Any unsatisfactory automatic segmentation is corrected interactively with mouse clicks. We demonstrate the above segmentation procedure on a flower image database of 1078 samples from 113 species. 257 out of 1078 images are segmented without interactive correction. On average, 5.7 mouse clicks are needed for each image, and the segmentation process takes 15.2 seconds
Keywords :
computer vision; image segmentation; visual databases; flower image database; global parametric shape model; model-based interactive object segmentation procedure; parametric boundary; unsatisfactory automatic segmentation; Computer vision; Humans; Image databases; Image segmentation; Lighting; Mice; Object segmentation; Pixel; Shape; Visual databases;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Application of Computer Vision, 2005. WACV/MOTIONS '05 Volume 1. Seventh IEEE Workshops on
Conference_Location :
Breckenridge, CO
Print_ISBN :
0-7695-2271-8
Type :
conf
DOI :
10.1109/ACVMOT.2005.8
Filename :
4129527
Link To Document :
بازگشت