Deep Self-Taught Learning for Weakly Supervised Object Localization

TL;DR

This paper introduces a deep self-taught learning framework for weakly supervised object localization that iteratively improves detector accuracy by self-retraining with reliably identified positive samples, leading to superior performance.

Contribution

It proposes a novel self-taught learning method with seed sample acquisition, online supportive sample harvesting, and score-based guidance to enhance weakly supervised object localization.

Findings

Outperforms state-of-the-art methods on PASCAL datasets

Effectively improves positive sample quality during training

Demonstrates significant localization accuracy gains

Abstract

Most existing weakly supervised localization (WSL) approaches learn detectors by finding positive bounding boxes based on features learned with image-level supervision. However, those features do not contain spatial location related information and usually provide poor-quality positive samples for training a detector. To overcome this issue, we propose a deep self-taught learning approach, which makes the detector learn the object-level features reliable for acquiring tight positive samples and afterwards re-train itself based on them. Consequently, the detector progressively improves its detection ability and localizes more informative positive samples. To implement such self-taught learning, we propose a seed sample acquisition method via image-to-object transferring and dense subgraph discovery to find reliable positive samples for initializing the detector. An online supportive sample harvesting scheme is further proposed to dynamically select the most confident tight positive samples and train the detector in a mutual boosting way. To prevent the detector from being trapped in poor optima due to overfitting, we propose a new relative improvement of predicted CNN scores for guiding the self-taught learning process. Extensive experiments on PASCAL 2007 and 2012 show that our approach outperforms the state-of-the-arts, strongly validating its effectiveness.