SWA Object Detection

Abstract

Do you want to improve 1.0 AP for your object detector without any inference cost and any change to your detector? Let us tell you such a recipe. It is surprisingly simple: train your detector for an extra 12 epochs using cyclical learning rates and then average these 12 checkpoints as your final detection model}. This potent recipe is inspired by Stochastic Weights Averaging (SWA), which is proposed in arXiv:1803.05407 for improving generalization in deep neural networks. We found it also very effective in object detection. In this technique report, we systematically investigate the effects of applying SWA to object detection as well as instance segmentation. Through extensive experiments, we discover the aforementioned workable policy of performing SWA in object detection, and we consistently achieve $\sim$1.0 AP improvement over various popular detectors on the challenging COCO benchmark, including Mask RCNN, Faster RCNN, RetinaNet, FCOS, YOLOv3 and VFNet. We hope this work will make more researchers in object detection know this technique and help them train better object detectors. Code is available at: https://github.com/hyz-xmaster/swa_object_detection .

arXiv comment9 pages; polished

References in corpus (4)

• Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
• YOLOv3: An Incremental Improvement
• MMDetection: Open MMLab Detection Toolbox and Benchmark
• VarifocalNet: An IoU-aware Dense Object Detector

Cited by in corpus (5)

• Rethinking Rotated Object Detection with Gaussian Wasserstein Distance Loss
• You Only Look One-level Feature
• Real-Time Anchor-Free Single-Stage 3D Detection with IoU-Awareness
• LUAI Challenge 2021 on Learning to Understand Aerial Images
• The Second Place Solution for ICCV2021 VIPriors Instance Segmentation Challenge