通过在线伪标签提升多数据集目标检测精度
Online Pseudo-Label Unified Object Detection for Multiple Datasets Training
- 用周期更新的教师模型生成未标注样本的伪标签
- 在COCO、Object365等数据集上超越现有最优方法
- 适合需要融合多个数据集的统一目标检测场景
统一目标检测（UOD）任务旨在通过在多个数据集上训练，实现所有合并类别的一体化检测，对全面目标检测场景具有重要意义。本文深入分析了跨数据集缺失标注问题，提出一种在线伪标签统一目标检测方案。该方法利用周期性更新的教师模型为各子数据集中的未标注对象生成伪标签，周期更新策略能更好确保教师模型达到局部最优，从而最大化伪标签质量。此外，我们调研了重叠区域建议框对边界框回归精度的影响，提出类别特定的边界框回归与伪标签RPN头，以提升区域提议网络（RPN）的召回率。在常用基准数据集（如COCO、Object365和OpenImages）上的实验结果表明，所提在线伪标签UOD方法在精度上优于现有SOTA方法。
原文摘要 · Abstract (English)
The Unified Object Detection (UOD) task aims to achieve object detection of all merged categories through training on multiple datasets, and is of great significance in comprehensive object detection scenarios. In this paper, we conduct a thorough analysis of the cross datasets missing annotations issue, and propose an Online Pseudo-Label Unified Object Detection scheme. Our method uses a periodically updated teacher model to generate pseudo-labels for the unlabelled objects in each sub-dataset. This periodical update strategy could better ensure that the accuracy of the teacher model reaches the local maxima and maximized the quality of pseudo-labels. In addition, we survey the influence of overlapped region proposals on the accuracy of box regression. We propose a category specific box regression and a pseudo-label RPN head to improve the recall rate of the Region Proposal Network (PRN). Our experimental results on common used benchmarks (\eg COCO, Object365 and OpenImages) indicates that our online pseudo-label UOD method achieves higher accuracy than existing SOTA methods.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。