Contour Detection and Hierarchical Image Segmentation

Authors:
Arbelaez, Pablo
Maire, Michael
Fowlkes, Charless
Malik, Jitendra
Technical Report Identifier: EECS-2010-17
February 16, 2010
EECS-2010-17.pdf

Abstract: This paper investigates two fundamental problems in computer vision: contour detection and image segmentation. We present state-of-the-art algorithms for both of these tasks. Our contour detector combines multiple local cues into a globalization framework based on spectral clustering. Our segmentation algorithm consists of generic machinery for transforming the output of any contour detector into a hierarchical region tree. In this manner, we reduce the problem of image segmentation to that of contour detection. Extensive experimental evaluation demonstrates that both our contour detection and segmentation methods significantly outperform competing algorithms. The automatically generated hierarchical segmentations can be interactively refined by userspecified annotations. Computation at multiple image resolutions provides a means of coupling our system to recognition applications.