用流模型提升小目标定位抗噪声能力，避免标注错误导致的过拟合。
Noise-Robust Tiny Object Localization with Flows
- 通过归一化流建模复杂误差分布，灵活捕捉标注噪声。
- 在AI-TOD数据集上使DINO基线提升1.2% AP，显著改善小目标检测。
- 适合标注质量差或小目标密集场景的检测任务使用。
尽管通用目标检测取得显著进展，小目标与正常尺度目标之间仍存在明显性能差距。我们发现小目标对标注噪声极为敏感，严格优化定位目标可能导致噪声过拟合。为此，提出基于流的小目标定位框架TOLF，利用归一化流实现灵活的误差建模和不确定性引导优化。该方法通过流模型捕捉非高斯的预测分布，在噪声监督下实现稳健学习。此外，不确定性感知梯度调制机制可抑制高不确定性、易受噪声影响样本的学习，从而缓解过拟合并稳定训练过程。在三个数据集上的大量实验验证了该方法的有效性，尤其在AI-TOD数据集上，使DINO基线提升1.2% AP。
原文摘要 · Abstract (English)
Despite significant advances in generic object detection, a persistent performance gap remains for tiny objects compared to normal-scale objects. We demonstrate that tiny objects are highly sensitive to annotation noise, where optimizing strict localization objectives risks noise overfitting. To address this, we propose Tiny Object Localization with Flows (TOLF), a noise-robust localization framework leveraging normalizing flows for flexible error modeling and uncertainty-guided optimization. Our method captures complex, non-Gaussian prediction distributions through flow-based error modeling, enabling robust learning under noisy supervision. An uncertainty-aware gradient modulation mechanism further suppresses learning from high-uncertainty, noise-prone samples, mitigating overfitting while stabilizing training. Extensive experiments across three datasets validate our approach's effectiveness. Especially, TOLF boosts the DINO baseline by 1.2% AP on the AI-TOD 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。