Towards Noise-resistant Object Detection with Noisy Annotations

Abstract

Training deep object detectors requires significant amount of human-annotated images with accurate object labels and bounding box coordinates, which are extremely expensive to acquire. Noisy annotations are much more easily accessible, but they could be detrimental for learning. We address the challenging problem of training object detectors with noisy annotations, where the noise contains a mixture of label noise and bounding box noise. We propose a learning framework which jointly optimizes object labels, bounding box coordinates, and model parameters by performing alternating noise correction and model training. To disentangle label noise and bounding box noise, we propose a two-step noise correction method. The first step performs class-agnostic bounding box correction by minimizing classifier discrepancy and maximizing region objectness. The second step distils knowledge from dual detection heads for soft label correction and class-specific bounding box refinement. We conduct experiments on PASCAL VOC and MS-COCO dataset with both synthetic noise and machine-generated noise. Our method achieves state-of-the-art performance by effectively cleaning both label noise and bounding box noise. Code to reproduce all results will be released.

References in corpus (9)

• Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
• Mean teachers are better role models: Weight-averaged consistency targets improve semi-supervised deep learning results
• MMDetection: Open MMLab Detection Toolbox and Benchmark
• Weakly Supervised Object Localization with Multi-fold Multiple Instance Learning
• Regularizing Neural Networks by Penalizing Confident Output Distributions
• Unsupervised Label Noise Modeling and Loss Correction
• LSDA: Large Scale Detection Through Adaptation
• How does Disagreement Help Generalization against Label Corruption?
• Understanding and Utilizing Deep Neural Networks Trained with Noisy Labels

Cited by in corpus (6)

• Alpha-IoU: A Family of Power Intersection over Union Losses for Bounding Box Regression
• Evaluating Multi-label Classifiers with Noisy Labels
• Augmentation Matters: A Mix-Paste Method for X-Ray Prohibited Item Detection under Noisy Annotations
• Noisy Annotation Refinement for Object Detection
• Multi-Label Gold Asymmetric Loss Correction with Single-Label Regulators
• Sum of Ranked Range Loss for Supervised Learning