Weakly Supervised Localization using Deep Feature Maps

TL;DR

This paper introduces a weakly supervised object localization method that uses deep convolutional features and beam search to detect multiple objects, significantly improving localization accuracy without requiring object-level annotations.

Contribution

It presents a novel weakly supervised localization algorithm leveraging classification networks and beam search, outperforming previous methods on standard datasets.

Findings

Achieved an 8-point increase in mAP scores over state-of-the-art methods.

Effectively localizes multiple objects using only image-level labels.

Utilizes deep convolutional features for spatial and semantic pattern recognition.

Abstract

Object localization is an important computer vision problem with a variety of applications. The lack of large scale object-level annotations and the relative abundance of image-level labels makes a compelling case for weak supervision in the object localization task. Deep Convolutional Neural Networks are a class of state-of-the-art methods for the related problem of object recognition. In this paper, we describe a novel object localization algorithm which uses classification networks trained on only image labels. This weakly supervised method leverages local spatial and semantic patterns captured in the convolutional layers of classification networks. We propose an efficient beam search based approach to detect and localize multiple objects in images. The proposed method significantly outperforms the state-of-the-art in standard object localization data-sets with a 8 point increase in mAP scores.