Review for NeurIPS paper: Restoring Negative Information in Few-Shot Object Detection

Weaknesses: (1) If the insight is that hard negatives are important, then a very simple baseline presents itself: why not take a simple object detector trained on the base classes, and simply finetune the detector head and bbox regressor head in the usual way on the novel classes? This would automatically use the badly localized examples. I am surprised that the authors did not include this baseline. YOLO-FR uses DarkNet-19, Meta-Det uses VGG16, Meta-RCNN uses ResNet-101 (but without FPN or DCN). It is unclear what this paper pretrains it on.