Learning from Noisy Anchors for One-stage Object Detection

Abstract

State-of-the-art object detectors rely on regressing and classifying an extensive list of possible anchors, which are divided into positive and negative samples based on their intersection-over-union (IoU) with corresponding groundtruth objects. Such a harsh split conditioned on IoU results in binary labels that are potentially noisy and challenging for training. In this paper, we propose to mitigate noise incurred by imperfect label assignment such that the contributions of anchors are dynamically determined by a carefully constructed cleanliness score associated with each anchor. Exploring outputs from both regression and classification branches, the cleanliness scores, estimated without incurring any additional computational overhead, are used not only as soft labels to supervise the training of the classification branch but also sample re-weighting factors for improved localization and classification accuracy. We conduct extensive experiments on COCO, and demonstrate, among other things, the proposed approach steadily improves RetinaNet by ~2% with various backbones.

arXiv commentCVPR 2020 camera ready

References in corpus (15)

• FCOS: Fully Convolutional One-Stage Object Detection
• MentorNet: Learning Data-Driven Curriculum for Very Deep Neural Networks on Corrupted Labels
• Rich feature hierarchies for accurate object detection and semantic segmentation
• YOLO9000: Better, Faster, Stronger
• Deformable Convolutional Networks
• A Closer Look at Memorization in Deep Networks
• RepPoints: Point Set Representation for Object Detection
• Bottom-up Object Detection by Grouping Extreme and Center Points
• Cascade R-CNN: High Quality Object Detection and Instance Segmentation
• Scale-Aware Trident Networks for Object Detection
• Gradient Harmonized Single-stage Detector
• Learning from Noisy Labels with Distillation
• Region Proposal by Guided Anchoring
• Cascade RetinaNet: Maintaining Consistency for Single-Stage Object Detection
• Prime Sample Attention in Object Detection

Cited by in corpus (2)

• Generalized Focal Loss: Learning Qualified and Distributed Bounding Boxes for Dense Object Detection
• AutoAssign: Differentiable Label Assignment for Dense Object Detection