用生成伪标签和合成数据提升跨域小样本目标检测性能
GiPL: Generative augmented iterative Pseudo-Labeling for Cross-Domain Few-Shot Object Detection
- 迭代伪标签融合真值，充分挖掘稀疏支持集
- 用视觉语言模型生成多物体合成图像，缓解过拟合
- 在3个数据集上1/5/10-shot均显著优于现有方法
视觉-语言基础模型在跨域小样本目标检测（CD-FSOD）中展现出良好的零样本泛化能力。然而，在微调阶段面临两大挑战：因单实例标注稀疏导致支持集利用不足，以及在极少量目标域样本下严重过拟合。为此，本文提出GiPL，一种高效的双分支训练框架。第一分支设计了迭代伪标签自训练范式，对支持集进行零样本推理生成可靠伪标注，与真实标签融合后迭代优化模型，充分挖掘支持集数据；第二分支引入基于大视觉-语言模型的生成式数据增强流程，合成领域对齐、多对象标注图像，丰富训练样本并抑制过拟合。在三个具有挑战性的CD-FSOD数据集（RUOD、CARPK、CarDD）上，1/5/10-shot设置下的大量实验表明，GiPL持续优于当前最优方法，性能显著提升。代码已公开于CDiscover。
原文摘要 · Abstract (English)
Vision-language foundation models have shown promising zero-shot generalization for Cross-Domain Few-Shot Object Detection (CD-FSOD). However, they face two critical challenges in fine-tuning: insufficient support set utilization due to sparse single-instance annotations, and severe overfitting under extremely limited target-domain samples. To address these issues, this paper proposes GiPL, an efficient two-branch training framework. In the first branch, we design an iterative pseudo-label self-training paradigm, which performs zero-shot inference on the support set to generate reliable pseudo-annotations, fuses them with ground-truth labels, and iteratively optimizes the model to fully exploit support set data. In the second branch, we introduce generative data augmentation pipeline using large vision-language models, which synthesizes domain-aligned, multi-object annotated images to enrich training samples and suppress overfitting. Extensive experiments on three challenging CD-FSOD datasets (RUOD, CARPK, CarDD) under 1/5/10-shot settings demonstrate that GiPL consistently outperforms state-of-the-art methods with significant performance gains. Code is available at \href{https://github.com/z-yaz/CDiscover}{CDiscover}.
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。