Dynamic learning from multiple examples for semantic object segmentation and search
- 作者：Xu ; Yaowu ; Saber ; Eli ; Tekalp ; A. Murat
- 关键词：Learning by examples ; Dynamic learning ; Shape matching ; Segmentation
- 刊名：Computer Vision and Image Understanding
- 出版年：2004
- 期刊代码：125_10773142
- 类别：cp
- 出版时间：September, 2004
- 卷：95
- 期：3
- 页码：334-353
- 文件大小：5.90 M
摘要
We present a novel 芒芒芒dynamic learning芒芒芒 approach for an intelligent image database system to automatically improve object segmentation and labeling without user intervention, as new examples become available, for object-based indexing. The proposed approach is an extension of our earlier work on 芒芒芒learning by example,芒芒芒 which addressed labeling of similar objects in a set of database images based on a single example. The proposed dynamic learning procedure utilizes multiple example object templates to improve the accuracy of existing object segmentations and labels. Multiple example templates may be images of the same object from different viewing angles, or images of related objects. This paper also introduces a new shape similarity metric called normalized area of symmetric differences (NASD), which has desired properties for use in the proposed 芒芒芒dynamic learning芒芒芒 scheme, and is more robust against boundary noise that results from automatic image segmentation. Performance of the dynamic learning procedures has been demonstrated by experimental results.