Salvato in:
Dettagli Bibliografici
Autori principali: Wu, Ruiqi, Su, Na, Zhang, Chenran, Ma, Tengfei, Zhou, Tao, Cui, Zhiting, Tang, Nianfeng, Mao, Tianyu, Zhou, Yi, Fan, Wen, Wu, Tianxing, Jing, Shenqi, Fu, Huazhu
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:https://arxiv.org/abs/2501.15798
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866916912577904640
author Wu, Ruiqi
Su, Na
Zhang, Chenran
Ma, Tengfei
Zhou, Tao
Cui, Zhiting
Tang, Nianfeng
Mao, Tianyu
Zhou, Yi
Fan, Wen
Wu, Tianxing
Jing, Shenqi
Fu, Huazhu
author_facet Wu, Ruiqi
Su, Na
Zhang, Chenran
Ma, Tengfei
Zhou, Tao
Cui, Zhiting
Tang, Nianfeng
Mao, Tianyu
Zhou, Yi
Fan, Wen
Wu, Tianxing
Jing, Shenqi
Fu, Huazhu
contents Vision-language pretraining (VLP) has been investigated to generalize across diverse downstream tasks for fundus image analysis. Although recent methods showcase promising achievements, they significantly rely on large-scale private image-text data but pay less attention to the pretraining manner, which limits their further advancements. In this work, we introduce MM-Retinal V2, a high-quality image-text paired dataset comprising CFP, FFA, and OCT image modalities. Then, we propose a novel fundus vision-language pretraining model, namely KeepFIT V2, which is pretrained by integrating knowledge from the elite data spark into categorical public datasets. Specifically, a preliminary textual pretraining is adopted to equip the text encoder with primarily ophthalmic textual knowledge. Moreover, a hybrid image-text knowledge injection module is designed for knowledge transfer, which is essentially based on a combination of global semantic concepts from contrastive learning and local appearance details from generative learning. Extensive experiments across zero-shot, few-shot, and linear probing settings highlight the generalization and transferability of KeepFIT V2, delivering performance competitive to state-of-the-art fundus VLP models trained on large-scale private image-text datasets. Our dataset and model are publicly available via https://github.com/lxirich/MM-Retinal.
format Preprint
id arxiv_https___arxiv_org_abs_2501_15798
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle MM-Retinal V2: Transfer an Elite Knowledge Spark into Fundus Vision-Language Pretraining
Wu, Ruiqi
Su, Na
Zhang, Chenran
Ma, Tengfei
Zhou, Tao
Cui, Zhiting
Tang, Nianfeng
Mao, Tianyu
Zhou, Yi
Fan, Wen
Wu, Tianxing
Jing, Shenqi
Fu, Huazhu
Computer Vision and Pattern Recognition
Vision-language pretraining (VLP) has been investigated to generalize across diverse downstream tasks for fundus image analysis. Although recent methods showcase promising achievements, they significantly rely on large-scale private image-text data but pay less attention to the pretraining manner, which limits their further advancements. In this work, we introduce MM-Retinal V2, a high-quality image-text paired dataset comprising CFP, FFA, and OCT image modalities. Then, we propose a novel fundus vision-language pretraining model, namely KeepFIT V2, which is pretrained by integrating knowledge from the elite data spark into categorical public datasets. Specifically, a preliminary textual pretraining is adopted to equip the text encoder with primarily ophthalmic textual knowledge. Moreover, a hybrid image-text knowledge injection module is designed for knowledge transfer, which is essentially based on a combination of global semantic concepts from contrastive learning and local appearance details from generative learning. Extensive experiments across zero-shot, few-shot, and linear probing settings highlight the generalization and transferability of KeepFIT V2, delivering performance competitive to state-of-the-art fundus VLP models trained on large-scale private image-text datasets. Our dataset and model are publicly available via https://github.com/lxirich/MM-Retinal.
title MM-Retinal V2: Transfer an Elite Knowledge Spark into Fundus Vision-Language Pretraining
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2501.15798