Saved in:
Bibliographic Details
Main Authors: Duan, Yue, Gu, Zhangxuan, Ying, Zhenzhe, Qi, Lei, Meng, Changhua, Shi, Yinghuan
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2408.01349
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914897886969856
author Duan, Yue
Gu, Zhangxuan
Ying, Zhenzhe
Qi, Lei
Meng, Changhua
Shi, Yinghuan
author_facet Duan, Yue
Gu, Zhangxuan
Ying, Zhenzhe
Qi, Lei
Meng, Changhua
Shi, Yinghuan
contents In the realm of cross-modal retrieval, seamlessly integrating diverse modalities within multimedia remains a formidable challenge, especially given the complexities introduced by noisy correspondence learning (NCL). Such noise often stems from mismatched data pairs, which is a significant obstacle distinct from traditional noisy labels. This paper introduces Pseudo-Classification based Pseudo-Captioning (PC$^2$) framework to address this challenge. PC$^2$ offers a threefold strategy: firstly, it establishes an auxiliary "pseudo-classification" task that interprets captions as categorical labels, steering the model to learn image-text semantic similarity through a non-contrastive mechanism. Secondly, unlike prevailing margin-based techniques, capitalizing on PC$^2$'s pseudo-classification capability, we generate pseudo-captions to provide more informative and tangible supervision for each mismatched pair. Thirdly, the oscillation of pseudo-classification is borrowed to assistant the correction of correspondence. In addition to technical contributions, we develop a realistic NCL dataset called Noise of Web (NoW), which could be a new powerful NCL benchmark where noise exists naturally. Empirical evaluations of PC$^2$ showcase marked improvements over existing state-of-the-art robust cross-modal retrieval techniques on both simulated and realistic datasets with various NCL settings. The contributed dataset and source code are released at https://github.com/alipay/PC2-NoiseofWeb.
format Preprint
id arxiv_https___arxiv_org_abs_2408_01349
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle PC$^2$: Pseudo-Classification Based Pseudo-Captioning for Noisy Correspondence Learning in Cross-Modal Retrieval
Duan, Yue
Gu, Zhangxuan
Ying, Zhenzhe
Qi, Lei
Meng, Changhua
Shi, Yinghuan
Multimedia
Artificial Intelligence
Computer Vision and Pattern Recognition
Information Retrieval
Machine Learning
In the realm of cross-modal retrieval, seamlessly integrating diverse modalities within multimedia remains a formidable challenge, especially given the complexities introduced by noisy correspondence learning (NCL). Such noise often stems from mismatched data pairs, which is a significant obstacle distinct from traditional noisy labels. This paper introduces Pseudo-Classification based Pseudo-Captioning (PC$^2$) framework to address this challenge. PC$^2$ offers a threefold strategy: firstly, it establishes an auxiliary "pseudo-classification" task that interprets captions as categorical labels, steering the model to learn image-text semantic similarity through a non-contrastive mechanism. Secondly, unlike prevailing margin-based techniques, capitalizing on PC$^2$'s pseudo-classification capability, we generate pseudo-captions to provide more informative and tangible supervision for each mismatched pair. Thirdly, the oscillation of pseudo-classification is borrowed to assistant the correction of correspondence. In addition to technical contributions, we develop a realistic NCL dataset called Noise of Web (NoW), which could be a new powerful NCL benchmark where noise exists naturally. Empirical evaluations of PC$^2$ showcase marked improvements over existing state-of-the-art robust cross-modal retrieval techniques on both simulated and realistic datasets with various NCL settings. The contributed dataset and source code are released at https://github.com/alipay/PC2-NoiseofWeb.
title PC$^2$: Pseudo-Classification Based Pseudo-Captioning for Noisy Correspondence Learning in Cross-Modal Retrieval
topic Multimedia
Artificial Intelligence
Computer Vision and Pattern Recognition
Information Retrieval
Machine Learning
url https://arxiv.org/abs/2408.01349