ExDet: Open-Domain Open-Vocabulary Detection with Cross-modal Extrapolation and Rectification
2026-06-08 • Computer Vision and Pattern Recognition
Computer Vision and Pattern Recognition
AI summaryⓘ
The authors propose ExDet, a new method to help object detectors recognize new types of objects in new environments without needing expensive retraining. Their approach uses a few smart components to adjust how detectors understand categories and domains by leveraging text descriptions and correcting the detector's outputs during inference. This helps improve detection especially when objects and backgrounds look different from what the detector saw during training. The method works better on several challenging datasets compared to previous approaches.
open-vocabulary detectiondomain generalizationvision-language modelsobject detectionprototype learningregion proposal networkrepresentation rectificationcross-domaincross-category
Authors
Yupeng Zhang, Yuzhong Feng, Ruize Han, Zhiwei Chen, Wei Feng, Liang Wan
Abstract
Open-domain open-vocabulary detection (ODOVD) requires detectors to generalize to both novel categories and unseen domains, making it more challenging than open-vocabulary detection. Existing methods typically train open-vocabulary detectors together with domain generalization modules from scratch, leading to high training cost. we propose ExDet, a lightweight category-domain collaborative generalization framework for ODOVD that enhances the cross-category and cross-domain generalization of existing detectors. ExDet consists of Text-Guided Extrapolation (TGE), a lightweight Detector-Compatible Rectification (DCR) module, and ExRPN. Specifically, TGE exploits the DeltaSpace property of vision-language models (VLMs) to infer category- and domain-aware proxy visual prototypes from text. DCR is learned from the TGE-generated prototypes in a detector training-free and real-data-free manner, and is inserted after the classification head at inference to rectify representations toward a detector-compatible source-domain visual distribution, thereby enhancing classification for targets from novel categories and unseen domains. ExRPN recalibrates proposal scores by combining semantic similarity with RPN confidence, improving recall for novel and domain-shifted objects while providing better support for subsequent classification and DCR. ExDet achieves SOTA performance on OD-LVIS, OV-LVIS, Objects365, and MSOSB.