Selective Search
Selective search is a region proposal method for object detection. It hierarchically groups similar regions based on color, texture, size, and shape.
Selective Search uses over-segmented images as input. Then takes the following steps.
1. Add all bounding boxes corresponding to segment parts to the region proposal list.
2. Group adjacent segments based on similarity
3. Go back to step 1 and repeat.
It is a “hierarchically” grouping because it is generating region proposals from small segments to large segments, taking a bottom-up approach.
