Directional Edge Boxes: Exploiting Inner Normal Direction Cues for Effective Object Proposal Generation
One of my papers, Directional Edge Boxes: Exploiting Inner Normal Direction Cues for Effective Object Proposal Generation is published recently on Journal of Computer Science and Technology, July 2017, Volume 32, Issue 4, pp 701–713, a special issue on deep learning.
Edges are important cues for localizing object proposals. The recent progresses to this problem are mostly driven by defining effective objectness measures based on edge cues.
In this paper, we develop a new representation named directional edges on which each edge pixel is assigned with a direction toward object center, through learning a direction prediction model with convolutional neural networks in a holistic manner. Based on directional edges, two new objectness measures are designed for ranking object proposals.
Experiments show that the proposed method achieves 97.1% object recall at an overlap threshold of 0.5 and 81.9% object recall at an overlap threshold of 0.7 at 1000 proposals on the PASCAL VOC 2007 test dataset, which is superior to the state-of-the-art methods.
convolutional neural network
View the full PDF version here.