Loading [MathJax]/extensions/MathMenu.js
USER: Unified Semantic Enhancement With Momentum Contrast for Image-Text Retrieval | IEEE Journals & Magazine | IEEE Xplore

USER: Unified Semantic Enhancement With Momentum Contrast for Image-Text Retrieval


Abstract:

As a fundamental and challenging task in bridging language and vision domains, Image-Text Retrieval (ITR) aims at searching for the target instances that are semantically...Show More

Abstract:

As a fundamental and challenging task in bridging language and vision domains, Image-Text Retrieval (ITR) aims at searching for the target instances that are semantically relevant to the given query from the other modality, and its key challenge is to measure the semantic similarity across different modalities. Although significant progress has been achieved, existing approaches typically suffer from two major limitations: (1) It hurts the accuracy of the representation by directly exploiting the bottom-up attention based region-level features where each region is equally treated. (2) It limits the scale of negative sample pairs by employing the mini-batch based end-to-end training mechanism. To address these limitations, we propose a Unified Semantic Enhancement Momentum Contrastive Learning (USER) method for ITR. Specifically, we delicately design two simple but effective Global representation based Semantic Enhancement (GSE) modules. One learns the global representation via the self-attention algorithm, noted as Self-Guided Enhancement (SGE) module. The other module benefits from the pre-trained CLIP module, which provides a novel scheme to exploit and transfer the knowledge from an off-the-shelf model, noted as CLIP-Guided Enhancement (CGE) module. Moreover, we incorporate the training mechanism of MoCo into ITR, in which two dynamic queues are employed to enrich and enlarge the scale of negative sample pairs. Meanwhile, a Unified Training Objective (UTO) is developed to learn from mini-batch based and dynamic queue based samples. Extensive experiments on the benchmark MSCOCO and Flickr30K datasets demonstrate the superiority of both retrieval accuracy and inference efficiency. For instance, compared with the existing best method NAAF, the metric R@1 of our USER on the MSCOCO 5K Testing set is improved by 5% and 2.4% on caption retrieval and image retrieval without any external knowledge or pre-trained model while enjoying over 60 times faster inference speed. O...
Published in: IEEE Transactions on Image Processing ( Volume: 33)
Page(s): 595 - 609
Date of Publication: 05 January 2024

ISSN Information:

PubMed ID: 38190676

Funding Agency:

Author image of Yan Zhang
Tianjin Key Laboratory of Brain-Inspired Intelligence Technology, School of Electrical and Information Engineering, Tianjin University, Tianjin, China
Yan Zhang received the M.S. degree in control engineering from Tianjin University, Tianjin, China, in 2021, where he is currently pursuing the Ph.D. degree with the School of Electrical and Information Engineering. His current research interests include cross-modal retrieval, multi-modal analysis, and computer vision.
Yan Zhang received the M.S. degree in control engineering from Tianjin University, Tianjin, China, in 2021, where he is currently pursuing the Ph.D. degree with the School of Electrical and Information Engineering. His current research interests include cross-modal retrieval, multi-modal analysis, and computer vision.View more
Author image of Zhong Ji
Tianjin Key Laboratory of Brain-Inspired Intelligence Technology, School of Electrical and Information Engineering, Tianjin University, Tianjin, China
Shanghai Artificial Intelligence Laboratory, Shanghai, China
Zhong Ji (Senior Member, IEEE) received the Ph.D. degree in signal and information processing from Tianjin University, Tianjin, China, in 2008. He is currently a Professor with the School of Electrical and Information Engineering, Tianjin University. He has authored over 100 scientific articles in refereed journals and proceedings. His current research interests include multimedia understanding, few-shot learning, cross-m...Show More
Zhong Ji (Senior Member, IEEE) received the Ph.D. degree in signal and information processing from Tianjin University, Tianjin, China, in 2008. He is currently a Professor with the School of Electrical and Information Engineering, Tianjin University. He has authored over 100 scientific articles in refereed journals and proceedings. His current research interests include multimedia understanding, few-shot learning, cross-m...View more
Author image of Di Wang
Tianjin Key Laboratory of Brain-Inspired Intelligence Technology, School of Electrical and Information Engineering, Tianjin University, Tianjin, China
Di Wang received the Ph.D. degree from Tianjin University, China, in 2018. He is currently a Lecturer with the School of Electrical and Information Engineering, Tianjin University. His current research interests include conformal prediction and machine learning.
Di Wang received the Ph.D. degree from Tianjin University, China, in 2018. He is currently a Lecturer with the School of Electrical and Information Engineering, Tianjin University. His current research interests include conformal prediction and machine learning.View more
Author image of Yanwei Pang
Tianjin Key Laboratory of Brain-Inspired Intelligence Technology, School of Electrical and Information Engineering, Tianjin University, Tianjin, China
Shanghai Artificial Intelligence Laboratory, Shanghai, China
Yanwei Pang (Senior Member, IEEE) received the Ph.D. degree in electronic engineering from the University of Science and Technology of China, Hefei, China, in 2004. He is currently a Professor with the School of Electronic Information Engineering, Tianjin University, Tianjin, China. He has authored over 150 scientific articles. His current research interests include machine learning, MRI, and computer vision.
Yanwei Pang (Senior Member, IEEE) received the Ph.D. degree in electronic engineering from the University of Science and Technology of China, Hefei, China, in 2004. He is currently a Professor with the School of Electronic Information Engineering, Tianjin University, Tianjin, China. He has authored over 150 scientific articles. His current research interests include machine learning, MRI, and computer vision.View more
Key Laboratory of Intelligent Interaction and Applications, Ministry of Industry and Information Technology, and the School of Artificial Intelligence, OPtics and ElectroNics (iOPEN), Northwestern Polytechnical University, Xi’an, China
Xuelong Li (Fellow, IEEE) is currently a Full Professor with the School of Artificial Intelligence, OPtics and ElectroNics (iOPEN), Northwestern Polytechnical University, Xi’an, China.
Xuelong Li (Fellow, IEEE) is currently a Full Professor with the School of Artificial Intelligence, OPtics and ElectroNics (iOPEN), Northwestern Polytechnical University, Xi’an, China.View more

Author image of Yan Zhang
Tianjin Key Laboratory of Brain-Inspired Intelligence Technology, School of Electrical and Information Engineering, Tianjin University, Tianjin, China
Yan Zhang received the M.S. degree in control engineering from Tianjin University, Tianjin, China, in 2021, where he is currently pursuing the Ph.D. degree with the School of Electrical and Information Engineering. His current research interests include cross-modal retrieval, multi-modal analysis, and computer vision.
Yan Zhang received the M.S. degree in control engineering from Tianjin University, Tianjin, China, in 2021, where he is currently pursuing the Ph.D. degree with the School of Electrical and Information Engineering. His current research interests include cross-modal retrieval, multi-modal analysis, and computer vision.View more
Author image of Zhong Ji
Tianjin Key Laboratory of Brain-Inspired Intelligence Technology, School of Electrical and Information Engineering, Tianjin University, Tianjin, China
Shanghai Artificial Intelligence Laboratory, Shanghai, China
Zhong Ji (Senior Member, IEEE) received the Ph.D. degree in signal and information processing from Tianjin University, Tianjin, China, in 2008. He is currently a Professor with the School of Electrical and Information Engineering, Tianjin University. He has authored over 100 scientific articles in refereed journals and proceedings. His current research interests include multimedia understanding, few-shot learning, cross-modal analysis, and continual learning.
Zhong Ji (Senior Member, IEEE) received the Ph.D. degree in signal and information processing from Tianjin University, Tianjin, China, in 2008. He is currently a Professor with the School of Electrical and Information Engineering, Tianjin University. He has authored over 100 scientific articles in refereed journals and proceedings. His current research interests include multimedia understanding, few-shot learning, cross-modal analysis, and continual learning.View more
Author image of Di Wang
Tianjin Key Laboratory of Brain-Inspired Intelligence Technology, School of Electrical and Information Engineering, Tianjin University, Tianjin, China
Di Wang received the Ph.D. degree from Tianjin University, China, in 2018. He is currently a Lecturer with the School of Electrical and Information Engineering, Tianjin University. His current research interests include conformal prediction and machine learning.
Di Wang received the Ph.D. degree from Tianjin University, China, in 2018. He is currently a Lecturer with the School of Electrical and Information Engineering, Tianjin University. His current research interests include conformal prediction and machine learning.View more
Author image of Yanwei Pang
Tianjin Key Laboratory of Brain-Inspired Intelligence Technology, School of Electrical and Information Engineering, Tianjin University, Tianjin, China
Shanghai Artificial Intelligence Laboratory, Shanghai, China
Yanwei Pang (Senior Member, IEEE) received the Ph.D. degree in electronic engineering from the University of Science and Technology of China, Hefei, China, in 2004. He is currently a Professor with the School of Electronic Information Engineering, Tianjin University, Tianjin, China. He has authored over 150 scientific articles. His current research interests include machine learning, MRI, and computer vision.
Yanwei Pang (Senior Member, IEEE) received the Ph.D. degree in electronic engineering from the University of Science and Technology of China, Hefei, China, in 2004. He is currently a Professor with the School of Electronic Information Engineering, Tianjin University, Tianjin, China. He has authored over 150 scientific articles. His current research interests include machine learning, MRI, and computer vision.View more
Key Laboratory of Intelligent Interaction and Applications, Ministry of Industry and Information Technology, and the School of Artificial Intelligence, OPtics and ElectroNics (iOPEN), Northwestern Polytechnical University, Xi’an, China
Xuelong Li (Fellow, IEEE) is currently a Full Professor with the School of Artificial Intelligence, OPtics and ElectroNics (iOPEN), Northwestern Polytechnical University, Xi’an, China.
Xuelong Li (Fellow, IEEE) is currently a Full Professor with the School of Artificial Intelligence, OPtics and ElectroNics (iOPEN), Northwestern Polytechnical University, Xi’an, China.View more

Contact IEEE to Subscribe

References

References is not available for this document.