Robust Object Detection With Inaccurate Bounding Boxes

TL;DR

This paper introduces OA-MIL, a novel approach that improves object detection robustness against inaccurate bounding boxes by leveraging classification guidance and instance selection, validated on noisy datasets.

Contribution

Proposes OA-MIL, an object-aware multiple instance learning method that refines localization using classification signals and instance selection to handle noisy bounding boxes.

Findings

Effective on synthetic noisy datasets like PASCAL VOC and MS-COCO.

Improves localization accuracy despite inaccurate bounding box annotations.

Demonstrates robustness on real noisy wheat head dataset.

Abstract

Learning accurate object detectors often requires large-scale training data with precise object bounding boxes. However, labeling such data is expensive and time-consuming. As the crowd-sourcing labeling process and the ambiguities of the objects may raise noisy bounding box annotations, the object detectors will suffer from the degenerated training data. In this work, we aim to address the challenge of learning robust object detectors with inaccurate bounding boxes. Inspired by the fact that localization precision suffers significantly from inaccurate bounding boxes while classification accuracy is less affected, we propose leveraging classification as a guidance signal for refining localization results. Specifically, by treating an object as a bag of instances, we introduce an Object-Aware Multiple Instance Learning approach (OA-MIL), featured with object-aware instance selection and object-aware instance extension. The former aims to select accurate instances for training, instead of directly using inaccurate box annotations. The latter focuses on generating high-quality instances for selection. Extensive experiments on synthetic noisy datasets (i.e., noisy PASCAL VOC and MS-COCO) and a real noisy wheat head dataset demonstrate the effectiveness of our OA-MIL. Code is available at https://github.com/cxliu0/OA-MIL.