Dual Relation Knowledge Distillation for Object Detection

TL;DR

This paper introduces dual relation knowledge distillation (DRKD), a novel method that enhances object detection models by capturing global pixel relations and instance similarities, addressing feature imbalance and small object representation issues.

Contribution

The paper proposes a new distillation approach combining pixel-wise and instance-wise relations, improving detection performance for both one-stage and two-stage detectors.

Findings

Achieves state-of-the-art mAP improvements on COCO 2017.

Effectively addresses foreground-background feature imbalance.

Enhances small object detection through relation distillation.

Abstract

Knowledge distillation is an effective method for model compression. However, it is still a challenging topic to apply knowledge distillation to detection tasks. There are two key points resulting in poor distillation performance for detection tasks. One is the serious imbalance between foreground and background features, another one is that small object lacks enough feature representation. To solve the above issues, we propose a new distillation method named dual relation knowledge distillation (DRKD), including pixel-wise relation distillation and instance-wise relation distillation. The pixel-wise relation distillation embeds pixel-wise features in the graph space and applies graph convolution to capture the global pixel relation. By distilling the global pixel relation, the student detector can learn the relation between foreground and background features, and avoid the difficulty of distilling features directly for the feature imbalance issue. Besides, we find that instance-wise relation supplements valuable knowledge beyond independent features for small objects. Thus, the instance-wise relation distillation is designed, which calculates the similarity of different instances to obtain a relation matrix. More importantly, a relation filter module is designed to highlight valuable instance relations. The proposed dual relation knowledge distillation is general and can be easily applied for both one-stage and two-stage detectors. Our method achieves state-of-the-art performance, which improves Faster R-CNN based on ResNet50 from 38.4% to 41.6% mAP and improves RetinaNet based on ResNet50 from 37.4% to 40.3% mAP on COCO 2017.