Contour Detection and Hierarchical Image Segmentation

TLDR

The paper investigates contour detection and image segmentation in computer vision. The authors propose state‑of‑the‑art algorithms that fuse local cues via spectral clustering for contour detection and transform any contour detector output into a hierarchical region tree, thereby reducing segmentation to contour detection and enabling multi‑resolution computation for recognition. Experiments show the proposed methods outperform existing ones, and the resulting hierarchical segmentations can be interactively refined by user annotations.

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 user-specified annotations. Computation at multiple image resolutions provides a means of coupling our system to recognition applications.