CDFSOD-benchmark is a research project focusing on cross-domain few-shot object detection (CD-FSOD). It was released by the research team of Beijing University of Aeronautics and Astronautics and iFlytek in 2024. The related paper results are "Cross-Domain Few-Shot Object Detection via Enhanced Open-Set Object Detector", has been accepted by ECCV24. This project aims to solve the problem of small-sample object detection when there is a significant domain difference between the source domain and the target domain. It includes a dataset for algorithm evaluation, as well as style, inter-class variance (ICV), and indefinable boundaries (IB) for measuring domain differences.

Citation

@inproceedings{fu2025cross, title={Cross-domain few-shot object detection via enhanced open-set object detector}, author={Fu, Yuqian and Wang, Yu and Pan, Yixuan and Huai, Lian and Qiu, Xingyu and Shangguan, Zeyu and Liu, Tong and Fu, Yanwei and Van Gool, Luc and Jiang, Xingqun} booktitle={European Conference on Computer Vision}, pages={247–264}, year={2025}, organization={Springer} } @inproceedings{fu2025ntire, title={NTIRE 2025 challenge on cross-domain few-shot object detection: Methods and results}, Authors: {Fu, Yuqian and Qiu, Xingyu and Ren, Bin and Fu, Yanwei and Timofte, Radu and Sebe, Nicu and Yang, Ming-Hsuan and Van Gool, Luc and Zhang, Kaijin and Nong, Qingpeng and others} booktitle={Proceedings of the Computer Vision and Pattern Recognition Conference}, pages={1048–1069}, year={2025} }