Spatial reasoning for few-shot object detection

作者：

Highlights：

• We consider few-shot object detection that requires only a few training examples to detect novel categories.

• Inspired by a human visual system, we propose a spatial reasoning process to detect novel categories in a context that is less considered in few-shot object detection.

• To overcome a few-shot environment itself, we further present a spatial data augmentation method that efficiently enhance the ability of the spatial reasoning process.

• The proposed method significantly outperforms existing few-shot object detectors on the widely used PASCAL VOC and MS COCO datasets.

摘要

•We consider few-shot object detection that requires only a few training examples to detect novel categories.•Inspired by a human visual system, we propose a spatial reasoning process to detect novel categories in a context that is less considered in few-shot object detection.•To overcome a few-shot environment itself, we further present a spatial data augmentation method that efficiently enhance the ability of the spatial reasoning process.•The proposed method significantly outperforms existing few-shot object detectors on the widely used PASCAL VOC and MS COCO datasets.

论文关键词：Few-shot learning,Object detection,Transfer learning,Visual reasoning,Data augmentation

论文评审过程：Received 15 January 2021, Revised 14 May 2021, Accepted 14 June 2021, Available online 24 June 2021, Version of Record 2 July 2021.

论文官网地址：https://doi.org/10.1016/j.patcog.2021.108118