Detect Everything with Few Examples

arXiv:2309.12969v3 Announce Type: replace
Abstract: Few-shot object detection aims at detecting novel categories given a few example images. Recent methods focus on finetuning strategies, with complicated procedures that prohibit a wider application. In this paper, we introduce DE-ViT, a few-shot object detector without the need for finetuning. DE-ViT's novel architecture is based on a new region-propagation mechanism for localization. The propagated region masks are transformed into bounding boxes through a learnable spatial integral layer. Instead of training prototype classifiers, we …

abstract application architecture arxiv cs.cv detection everything example examples few-shot finetuning focus images novel object paper strategies type vit