Saliency Detection for Improving Object Proposals

TL;DR

This paper introduces a saliency detection method to refine object proposals, significantly enhancing their localization accuracy for better object detection performance.

Contribution

It presents a novel geodesic saliency detection technique applied to object proposals, improving their quality and ranking in detection tasks.

Findings

Significant improvement in proposal localization accuracy

Effective saliency-based refinement method

Enhanced detection performance on PASCAL VOC 2007

Abstract

Object proposals greatly benefit object detection task in recent state-of-the-art works. However, the existing object proposals usually have low localization accuracy at high intersection over union threshold. To address it, we apply saliency detection to each bounding box to improve their quality in this paper. We first present a geodesic saliency detection method in contour, which is designed to find closed contours. Then, we apply it to each candidate box with multi-sizes, and refined boxes can be easily produced in the obtained saliency maps which are further used to calculate saliency scores for proposal ranking. Experiments on PASCAL VOC 2007 test dataset demonstrate the proposed refinement approach can greatly improve existing models.