Performance-guided Reinforced Active Learning for Object Detection

TL;DR

This paper introduces MGRAL, a reinforcement learning-based active learning method for object detection that directly optimizes for mAP, reducing labeling effort while achieving superior performance on PASCAL VOC and COCO datasets.

Contribution

The paper proposes a novel mAP-guided reinforcement learning approach for active object detection, addressing batch selection complexity and computational efficiency.

Findings

MGRAL outperforms existing active learning strategies on PASCAL VOC and COCO.

It achieves the highest active learning curves with convincing visualizations.

The method effectively reduces labeling effort while maintaining high detection performance.

Abstract

Active learning (AL) strategies aim to train high-performance models with minimal labeling efforts, only selecting the most informative instances for annotation. Current approaches to evaluating data informativeness predominantly focus on the data's distribution or intrinsic information content and do not directly correlate with downstream task performance, such as mean average precision (mAP) in object detection. Thus, we propose Performance-guided (i.e. mAP-guided) Reinforced Active Learning for Object Detection (MGRAL), a novel approach that leverages the concept of expected model output changes as informativeness. To address the combinatorial explosion challenge of batch sample selection and the non-differentiable correlation between model performance and selected batches, MGRAL skillfully employs a reinforcement learning-based sampling agent that optimizes selection using policy gradient with mAP improvement as reward. Moreover, to reduce the computational overhead of mAP estimation with unlabeled samples, MGRAL utilizes an unsupervised way with fast look-up tables, ensuring feasible deployment. We evaluate MGRAL's active learning performance on detection tasks over PASCAL VOC and COCO benchmarks. Our approach demonstrates the highest AL curve with convincing visualizations, establishing a new paradigm in reinforcement learning-driven active object detection.