2022/10/11 by Wuti Xiong, Xiong, Wuti
Computer Science · #Domain Adaptation and Few-Shot Learning #Advanced Neural Network Applications #Multimodal Machine Learning Applications
paper · pdf · doi:10.48550/arxiv.2210.05311
In this paper, we propose a study of the cross-domain few-shot object detection (CD-FSOD) benchmark, consisting of image data from a diverse data domain. On the proposed benchmark, we evaluate state-of-art FSOD approaches, including meta-learning FSOD approaches and fine-tuning FSOD approaches. The results show that these methods tend to fall, and even underperform the naive fine-tuning model. We analyze the reasons for their failure and introduce a strong baseline that uses a mutually-beneficial manner to alleviate the overfitting problem. Our approach is remarkably superior to existing approaches by significant margins (2.0% on average) on the proposed benchmark. Our code is available at \urlhttps://github.com/FSOD/CD-FSOD.