Groups of adjacent contour segments for object detection
Abstract
We present a family of scale-invariant local shape features formed by chains of k connected, roughly straight contour segments (kAS), and their use for object class detection. kAS are able to cleanly encode pure fragments of an object boundary, without including nearby clutter. Moreover, they oer an attractive compromise between information content and repeatability, and encompass a wide variety of local shape structures.
Origin : Publisher files allowed on an open archive
Loading...