Feature reconstruction and metric based network for few-shot object detection

Yuewen Li, Wenquan Feng, Shuchang Lyu, Qi Zhao

Abstract: Highlights•We propose a meta-learning and metric-learning fused few-shot object detection model.•We propose meta-representation module to clearly cluster different categories.•We involve Pearson distance in metric-learning to decrease intra-class variance.•Experiments show that FM-FSOD achieves comparable results with previous SOTA methods.