Cross-domain Multi-modal Few-shot Object Detection via Rich Text

arXiv:2403.16188v1 Announce Type: new
Abstract: Cross-modal feature extraction and integration have led to steady performance improvements in few-shot learning tasks due to generating richer features. However, existing multi-modal object detection (MM-OD) methods degrade when facing significant domain-shift and are sample insufficient. We hypothesize that rich text information could more effectively help the model to build a knowledge relationship between the vision instance and its language description and can help mitigate domain shift. Specifically, we study the Cross-Domain few-shot generalization of …

abstract arxiv cs.cv detection domain extraction feature feature extraction features few-shot few-shot learning however improvements information integration modal multi-modal object performance sample shift tasks text type via