Towards Unsupervised Referring Expression Comprehension with Visual Semantic Parsing

Open Webpage

Yaodong Wang, Zhong Ji, Di Wang, Yanwei Pang, Xuelong Li

Published: 2024, Last Modified: 15 May 2025Knowl. Based Syst. 2024EveryoneRevisionsBibTeXCC BY-SA 4.0
Abstract: Highlights•Recognize the rich semantics contained in images.•Construct pseudo region-query pairs as training data.•Present a fine-grained multi-modality fusion decoder for REC.