Enhancing zero-shot object detection with external knowledge-guided robust contrast learning

Lijuan Duan, Guangyuan Liu, Qing En, Zhaoying Liu, Zhi Gong, Bian Ma

Abstract: Highlights•Using large language models to provide rich semantic information as external knowledge.•Supervised contrastive learning can optimize the distribution of visual features.•It is robust for both natural and fine-grained scenes.•Cycle consistency promotes that generated images maintain similar content and structure.