跳到正文
研踪PaperTrail
← 研踪 PaperTrail

arXiv:2606.19817 · v3

Training-Free Metrics for Synthetic Object Detection Data: A Proxy for Detector Performance

首次提交 2026-06-18 13:46
版本更新 2026-11-18 09:45
雷达观测 2026-11-19 19:08
arXiv ↗PDF ↗

摘要

Synthetic images are increasingly used to augment scarce real data for object detection. However, not all synthetic sets help equally, and the only way to know a set's value is to train a detector on it, which is slow and demands dense annotation. We ask whether a training-free metric can instead rank candidate synthetic training sets by their downstream utility. We propose Conditional-Composition Domain Match (CCDM), which converts any feature-space distance into a composition-stratified comparison, matching candidate and target within metadata-defined strata without training a detector. On COCO and VisDrone-DET, the best CCDM variant ranks 20 and 19 candidate training sets in strong agreement with YOLOv8 mAP (Spearman ρρ = 0.98 and 0.96), outperforming FID, KID, and MMD. Furthermore, CCDM holds when reference metadata comes from detector pseudo-labels instead of ground truth.

作者备注:19 pages, 12 figures

AI 论文解读

基于摘要 · 非全文审阅

此版本尚未生成 AI 解读。解读由后台任务生成并按论文版本保存,登录后共享已有结果。

版本差异

本站首次收录为 v3,暂无更早版本可对照。

观测到的版本

相邻记录可能跳过中间版本;首次发现 v2 及以上也会标为版本更新。

v3 · 2026-11-18 09:45 · 最新观测版本

Training-Free Metrics for Synthetic Object Detection Data: A Proxy for Detector Performance

Synthetic images are increasingly used to augment scarce real data for object detection. However, not all synthetic sets help equally, and the only way to know a set's value is to train a detector on it, which is slow and demands dense annotation. We ask whether a training-free metric can instead rank candidate synthetic training sets by their downstream utility. We propose Conditional-Composition Domain Match (CCDM), which converts any feature-space distance into a composition-stratified comparison, matching candidate and target within metadata-defined strata without training a detector. On COCO and VisDrone-DET, the best CCDM variant ranks 20 and 19 candidate training sets in strong agreement with YOLOv8 mAP (Spearman ρρ = 0.98 and 0.96), outperforming FID, KID, and MMD. Furthermore, CCDM holds when reference metadata comes from detector pseudo-labels instead of ground truth.

查看此版本解读 →查看此版本原文 ↗