解决检测模型对大目标偏倚问题，提升小目标检测效果
Unbiased Regression Loss for DETRs
- 提出尺寸归一化的L1损失，按宽高缩放每框贡献
- 在MS-COCO上实现大小目标检测均显著提升
- 适用于监督与半监督场景，尤其改善小目标表现
本文提出一种针对基于DETR检测器的新颖无偏回归损失。传统L1损失会偏向大目标，因其对总损失的贡献远超小目标，导致小目标检测性能下降。为此，新提出的尺寸L1损失（Sized L1 loss）根据每个目标的宽度和高度对框大小进行归一化。实验表明，在MS-COCO基准数据集上，该方法在全监督与半监督设置下均实现一致性能提升。
原文摘要 · Abstract (English)
In this paper, we introduce a novel unbiased regression loss for DETR-based detectors. The conventional $L_{1}$ regression loss tends to bias towards larger boxes, as they disproportionately contribute more towards the overall loss compared to smaller boxes. Consequently, the detection performance for small objects suffers. To alleviate this bias, the proposed new unbiased loss, termed Sized $L_{1}$ loss, normalizes the size of all boxes based on their individual width and height. Our experiments demonstrate consistent improvements in both fully-supervised and semi-supervised settings using the MS-COCO benchmark dataset.
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。