Rethinking Localization Map: Towards Accurate Object Perception with Self-Enhancement Maps

TL;DR

This paper introduces a new evaluation metric for weakly supervised object localization maps, proposes a self-enhancement method for more accurate localization using only category labels, and achieves state-of-the-art results on benchmark datasets.

Contribution

The work presents a direct pixel-wise evaluation metric for localization maps, annotates object masks for the dataset, and introduces a self-enhancement approach to improve localization accuracy with minimal supervision.

Findings

Achieved 54.88% localization accuracy on ILSVRC.

Proposed IoU-Threshold curves for better evaluation.

Developed a self-enhancement method that outperforms previous approaches.

Abstract

Recently, remarkable progress has been made in weakly supervised object localization (WSOL) to promote object localization maps. The common practice of evaluating these maps applies an indirect and coarse way, i.e., obtaining tight bounding boxes which can cover high-activation regions and calculating intersection-over-union (IoU) scores between the predicted and ground-truth boxes. This measurement can evaluate the ability of localization maps to some extent, but we argue that the maps should be measured directly and delicately, i.e., comparing the maps with the ground-truth object masks pixel-wisely. To fulfill the direct evaluation, we annotate pixel-level object masks on the ILSVRC validation set. We propose to use IoU-Threshold curves for evaluating the real quality of localization maps. Beyond the amended evaluation metric and annotated object masks, this work also introduces a novel self-enhancement method to harvest accurate object localization maps and object boundaries with only category labels as supervision. We propose a two-stage approach to generate the localization maps by simply comparing the similarity of point-wise features between the high-activation and the rest pixels. Based on the predicted localization maps, we explore to estimate object boundaries on a very large dataset. A hard-negative suppression loss is proposed for obtaining fine boundaries. We conduct extensive experiments on the ILSVRC and CUB benchmarks. In particular, the proposed Self-Enhancement Maps achieve the state-of-the-art localization accuracy of 54.88% on ILSVRC. The code and the annotated masks are released at https://github.com/xiaomengyc/SEM.