Extracting Image Regions by Structured Edge Prediction

We present two approaches to extract regions from struc-tured edge detection. While the state-of-the-art algorithm based on globalized probability of boundary (gPb) gener-ates a hierarchical region tree, it entails significant com-putational load. In this work, we exploit an efficient al-gorithm for structured edge prediction to extract regions. To generate high quality regions, we develop a novel al-gorithm to link the structured edge and gPb hierarchical image segmentation framework with steerable filters. The extracted regions are grouped by the proposed hierarchi-cal grouping method to generate object proposals for effec-tive detection and recognition problems. We demonstrate the effectiveness of our region generation for image seg-mentation on the BSDS500 database, and region generation for object proposals on the PASCAL VOC 2007 benchmark database. Experimental results show that the proposed al-gorithm achieves the comparable or superior quality to the state-of-the-art methods. 1