ExDet: Open-Domain Open-Vocabulary Detection with Cross-modal Extrapolation and Rectification
Researchers have introduced ExDet, a novel framework designed to improve open-domain open-vocabulary detection (ODOVD) capabilities. This lightweight system enhances the generalization of existing detectors to new categories and unseen domains without requiring training from scratch. ExDet utilizes text-guided extrapolation to infer visual prototypes and a detector-compatible rectification module to adjust representations, achieving state-of-the-art results on several benchmark datasets. AI
IMPACT Enhances generalization for object detection models, potentially improving performance in real-world applications with novel objects and diverse environments.