用于目标检测的双重关系知识蒸馏
Dual Relation Knowledge Distillation for Object Detection
AI总结：
针对目标检测中前景背景特征不平衡及小目标表示不足的问题，本文提出DRKD双重关系知识蒸馏方法，通过像素级与实例级关系蒸馏提升一阶段和两阶段检测器性能。
摘要：
知识蒸馏是一种有效的模型压缩方法。然而，将知识蒸馏应用于检测任务仍然是一个具有挑战性的课题。导致检测任务蒸馏性能较差的关键点有两个：一是前景与背景特征之间存在严重不平衡，二是小目标缺乏足够的特征表示。为解决上述问题，我们提出了一种名为双重关系知识蒸馏（DRKD）的新型蒸馏方法，包括像素级关系蒸馏和实例级关系蒸馏。像素级关系蒸馏将像素级特征嵌入到图空间中，并应用图卷积来捕获全局像素关系。通过蒸馏全局像素关系，学生检测器可以学习前景与背景特征之间的关系，并避免因特征不平衡问题而直接蒸馏特征的困难。此外，我们发现实例级关系为小目标补充了独立特征之外的宝贵知识。因此，设计了实例级关系蒸馏，通过计算不同实例的相似度来获得关系矩阵。更重要的是，设计了一个关系过滤模块以突出有价值的实例关系。所提出的双重关系知识蒸馏具有通用性，可以轻松应用于一阶段和两阶段检测器。我们的方法取得了最先进的性能，在 COCO 2017 上，将基于 ResNet50 的 Faster R-CNN 从 38.4% mAP 提升到 41.6% mAP，并将基于 ResNet50 的 RetinaNet 从 37.4% mAP 提升到 40.3% mAP。
英文摘要：
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.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
