Robust Few-Shot Aerial Image Object Detection via Unbiased Proposals Filtration

Few-shot aerial image object detection aims to rapidly detect object instances of novel category in aerial images by using few labeled samples. However, due to the complex background of aerial images, few labeled samples of novel categories, and the model trained with the few-shot learning paradigm is biased toward the base categories, it greatly increases the difficulty of identifying foreground objects of novel categories. In addition to this, tiny object detection is always a hot potato in aerial image object detection, and it is even more difficult for few-shot object detection. To this end, we propose a few-shot aerial image object detection with confidence-iou collaborative proposal filtration and tiny object constraint loss (FsCIT). Specifically, we first introduce a new confidence-iou collaborative proposal filtration scheme to region proposal network (RPN), which combines the unbiased intersection over union (IoU) scores between the two bounding boxes with foreground-background confidence scores to filter redundant region proposals and rescue more foreground proposals for the novel categories in RPN. Then, we design a new tiny object loss constraint term to attempt at overcoming the challenge of tiny object detection in few-shot aerial image object detection. This term considers the central point distance, the size of ground-truth bounding boxes, and the distances between the four edges of the ground-truth bounding box and the predicted bounding box. Experiments on object detection in optical remote sensing images (DIOR), tiny object detection for aerial images (AI-TOD), and high-resolution remote sensing detection (HRRSD) datasets show that FsCIT is effective and can improve the performance of few-shot aerial image object detection.