Experimental results of image-text retrieval on ECCV Caption.

<p>Experimental results of image-text retrieval on ECCV Caption.</p>

محفوظ في:

التفاصيل البيبلوغرافية
المؤلف الرئيسي:	Ran Jin (726426) (author)
مؤلفون آخرون:	Zhengang Li (714519) (author), Fang Deng (239069) (author), Yanhong Zhang (15518) (author), Min Luo (258201) (author), Tao Jin (55263) (author), Tengda Hou (22411238) (author), Chenjie Du (22411241) (author), Xiaozhe Gu (14097432) (author), Jie Yuan (208425) (author)
منشور في:	2025
الموضوعات:	Medicine Biotechnology Sociology Space Science Biological Sciences not elsewhere classified visual grounding capabilities standard datasets show significantly advanced cross overfitting problem caused method outperforms state issues significantly impede grained localization deficiency modal matching deficiency div >< p stage training framework stage framework grained alignment modal tasks modal fine second stage first stage thereby achieving text retrieval supplementary files spatial prompting recent years precise fine model ’ language pre label distillation hard labels extensive experiments contrastive relationships art approaches
الوسوم:	إضافة وسم لا توجد وسوم, كن أول من يضع وسما على هذه التسجيلة!

مواد مشابهة