Active Learning for Deep Object Detection via Probabilistic Modeling

TL;DR

This paper introduces a novel active learning method for deep object detection that estimates uncertainty using probabilistic modeling, reducing labeling costs while maintaining high performance.

Contribution

It proposes a probabilistic approach using mixture density networks to estimate uncertainties for localization and classification in object detection, improving efficiency.

Findings

Outperforms single-model active learning methods.

Comparable to multi-model methods at lower computational cost.

Effective on PASCAL VOC and MS-COCO datasets.

Abstract

Active learning aims to reduce labeling costs by selecting only the most informative samples on a dataset. Few existing works have addressed active learning for object detection. Most of these methods are based on multiple models or are straightforward extensions of classification methods, hence estimate an image's informativeness using only the classification head. In this paper, we propose a novel deep active learning approach for object detection. Our approach relies on mixture density networks that estimate a probabilistic distribution for each localization and classification head's output. We explicitly estimate the aleatoric and epistemic uncertainty in a single forward pass of a single model. Our method uses a scoring function that aggregates these two types of uncertainties for both heads to obtain every image's informativeness score. We demonstrate the efficacy of our approach in PASCAL VOC and MS-COCO datasets. Our approach outperforms single-model based methods and performs on par with multi-model based methods at a fraction of the computing cost.