Learning a Classification Model for Segmentation

We propose a two-class classification model for grouping. Human segmented natural images are used as positive examples. Negative examples of grouping are constructed by randomly matching human segmentations and images. In a preprocessing stage an image is oversegmented into superpixels. We define a variety of features derived from the classical Gestalt cues, including contour, texture, brightness and good continuation. Information-theoretic analysis is applied to evaluate the power of these grouping cues. We train a linear classifier to combine these features. To demonstrate the power of the classification model, a simple algorithm is used to randomly search for good segmentations. Results are shown on a wide range of images.

• Type: Conference
• Year: 2003
• Where: ICCV
• Authors: Xiaofeng Ren, Jitendra Malik
• Keywords: Classical Gestalt Cues, Classification Model, Computer Vision, Human Segmentations, ICCV 2003, Two-class Classification Model