Yafei Zhang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Yafei Zhang has a total of 76 co-authors. Here are the top co-authors by number of shared publications:
- Huafeng Li
Kunming University of Science and Technology
35 shared publications - Zhengtao Yu
Faculty of Information Engineering and Automation, Kunming University of Science and Technology, Kunming, China
16 shared publications - Dapeng Tao
Yunnan University
10 shared publications - Minghong Xie
Faculty of Information Engineering and Automation, Kunming University of Science and Technology, Kunming, China
6 shared publications - Yu Liu
Department of Biomedical Engineering, Hefei University of Technology, Hefei, China
6 shared publications Huafeng Li
Faculty of Information Engineering and Automation, Kunming University of Science and Technology, Kunming, China
4 shared publications
Latest Publications
- 10 June 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Emerging Topics in Computational Intelligence
- Vol. PP, 1-15
Cross-Modal Compensation and Dual Deformation Prediction Correction: Breaking Barriers in Unaligned Infrared-Visible Image Fusion
Unaligned infrared (IR)-visible (VIS) image fusion typically follows a two-stage paradigm: first registration, then fusion. However, this two-stage framework lacks simplicity and compactness compared to one-stage pattern, thereby increasing the complexity of model training and deployment. To overcome these obstacles, we propose a one-stage method for unaligned IR-VIS image fusion. Specifically, two pairs of aligned IR-VIS images are used to construct full-modal features without modality discrepancy, which serve as guidance for uni-modal feature extraction. With the designed Cross-Modal Information Compensation (CMIC), uni-modal image features are converted into full-modal features. In this way, not only the complementary information of IR-VIS images are preserved, but also the adverse impacts of modality differences on cross-modal feature alignment are effectively mitigated. Our method resolves the conflicting requirements of traditional cross-modal alignment and fusion methods, achieving simultaneous feature alignment and fusion within a one-stage scheme. Besides, a Dual Deformation Prediction and Correction (DDPC) mechanism is devised to enhance feature alignment accuracy. It leverages modality discrepancy-free features and gradient information to predict two deformation fields, which are then integrated to generate a refined deformation field for precise cross-modal feature alignment. As for effective aligned feature fusion, an Adaptive Perception Feature Fusion Network (APFFNet) is developed, which can adaptively perceive the importance of each fused feature, preventing excessive information loss during the fusion process. Compared with existing two-stage methods, the proposed approach achieves a superior balance between alignment and fusion. Additionally, our model is more compact and lightweight, promising more efficient training and deployment. Extensive experimental results validate the effectiveness of our proposed method.
- 10 June 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Emerging Topics in Computational Intelligence
- Vol. PP, 1-15
Cross-Modal Compensation and Dual Deformation Prediction Correction: Breaking Barriers in Unaligned Infrared-Visible Image Fusion
Unaligned infrared (IR)-visible (VIS) image fusion typically follows a two-stage paradigm: first registration, then fusion. However, this two-stage framework lacks simplicity and compactness compared to one-stage pattern, thereby increasing the complexity of model training and deployment. To overcome these obstacles, we propose a one-stage method for unaligned IR-VIS image fusion. Specifically, two pairs of aligned IR-VIS images are used to construct full-modal features without modality discrepancy, which serve as guidance for uni-modal feature extraction. With the designed Cross-Modal Information Compensation (CMIC), uni-modal image features are converted into full-modal features. In this way, not only the complementary information of IR-VIS images are preserved, but also the adverse impacts of modality differences on cross-modal feature alignment are effectively mitigated. Our method resolves the conflicting requirements of traditional cross-modal alignment and fusion methods, achieving simultaneous feature alignment and fusion within a one-stage scheme. Besides, a Dual Deformation Prediction and Correction (DDPC) mechanism is devised to enhance feature alignment accuracy. It leverages modality discrepancy-free features and gradient information to predict two deformation fields, which are then integrated to generate a refined deformation field for precise cross-modal feature alignment. As for effective aligned feature fusion, an Adaptive Perception Feature Fusion Network (APFFNet) is developed, which can adaptively perceive the importance of each fused feature, preventing excessive information loss during the fusion process. Compared with existing two-stage methods, the proposed approach achieves a superior balance between alignment and fusion. Additionally, our model is more compact and lightweight, promising more efficient training and deployment. Extensive experimental results validate the effectiveness of our proposed method.
- 2 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Image Processing
- Vol. 35  (10577149) , 1423-1435
Bidirectional Cross-Modal Collaborative Alignment via Semantic-Guided Visual Embeddings for Partially Relevant Video Retrieval
Partially Relevant Video Retrieval (PRVR) aims to retrieve videos that match a given textual query only partially. This task is inherently challenging due to the modality gap between text and video, which is further exacerbated by the partial semantic correspondence between linguistic descriptions and visual content. To address these challenges, we propose a bidirectional cross-modal alignment mechanism that collaboratively optimizes both visual and textual modalities. In the visual modality, a major difficulty lies in the absence of visual cues that directly correspond to textual semantics, limiting the models ability to align visual representations with textual meanings under unsupervised conditions. To overcome this issue, we construct a semantic-visual association library, which stores paired visual and textual features with semantic annotations. During training, the model dynamically retrieves the most semantically similar visual samples from this library based on the current visual feature vector. These retrieved samples, preliminarily associated with semantics via cross-modal matching, are used to form dynamic anchors that guide visual representation learning. By leveraging these enriched visual features, the model progressively refines the visual representations to achieve better alignment with the corresponding textual inputs, thereby enhancing cross-modal consistency. In the textual modality, we enhance textual representations by integrating semantically aligned visual features selected from the same association library, further narrowing the modality gap. Extensive experiments on benchmark datasets under partial semantic correspondence scenarios demonstrate that our method achieves state-of-the-art performance. The source code of the paper is available at https://github.com/cyanlll/BOA.
- 2 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Image Processing
- Vol. 35  (10577149) , 1423-1435
Bidirectional Cross-Modal Collaborative Alignment via Semantic-Guided Visual Embeddings for Partially Relevant Video Retrieval
Partially Relevant Video Retrieval (PRVR) aims to retrieve videos that match a given textual query only partially. This task is inherently challenging due to the modality gap between text and video, which is further exacerbated by the partial semantic correspondence between linguistic descriptions and visual content. To address these challenges, we propose a bidirectional cross-modal alignment mechanism that collaboratively optimizes both visual and textual modalities. In the visual modality, a major difficulty lies in the absence of visual cues that directly correspond to textual semantics, limiting the models ability to align visual representations with textual meanings under unsupervised conditions. To overcome this issue, we construct a semantic-visual association library, which stores paired visual and textual features with semantic annotations. During training, the model dynamically retrieves the most semantically similar visual samples from this library based on the current visual feature vector. These retrieved samples, preliminarily associated with semantics via cross-modal matching, are used to form dynamic anchors that guide visual representation learning. By leveraging these enriched visual features, the model progressively refines the visual representations to achieve better alignment with the corresponding textual inputs, thereby enhancing cross-modal consistency. In the textual modality, we enhance textual representations by integrating semantically aligned visual features selected from the same association library, further narrowing the modality gap. Extensive experiments on benchmark datasets under partial semantic correspondence scenarios demonstrate that our method achieves state-of-the-art performance. The source code of the paper is available at https://github.com/cyanlll/BOA.
- Top Cited Papers
- 10 January 2026
- Published by Elsevier in Pattern Recognition
- Vol. 175
Infrared-assisted single-stage framework for joint restoration and fusion of visible and infrared images under hazy conditions
- 10 January 2026
- Published by Elsevier in Pattern Recognition
- Vol. 175
Top Cited PapersInfrared-assisted single-stage framework for joint restoration and fusion of visible and infrared images under hazy conditions
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Image Processing
- Vol. 35, 8952-8965
Training-Free Video Corpus Moment Retrieval via Synergistic Collaboration and Adaptive Calibration
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Image Processing
- Vol. 35, 8952-8965
Training-Free Video Corpus Moment Retrieval via Synergistic Collaboration and Adaptive Calibration
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Image Processing
- Vol. 35, 7783-7798
Learning to Change by Critique and Correction: A Synergistic Framework for Remote Sensing Change Detection and Captioning
Change Detection (CD) and Change Captioning (CC) are two core tasks for understanding land-cover evolution in remote sensing imagery. Existing approaches have explored CD-CC joint modeling through shared representations, task-specific decoding, feature interaction, and semantic guidance. However, the lack of explicit cross-task feedback mechanisms often leads to mutual interference, making it difficult to achieve both accurate detection and expressive descriptions. To address this issue, we propose the Learning to Change by Critique and Correction (LCCC) framework, which reformulates CD and CC as a critique-correction closed-loop process. In LCCC, CD and CC no longer passively share features but interact through bidirectional critique and correction: the CD task provides explicit spatial constraints for CC, and CC, in turn, supervises CD via a Text-Guided Critique Attention (TGCA) mechanism, establishing a synergistic relationship where both tasks act as critics and correctors. Furthermore, we design a Reciprocal Suppression and Enhancement (RSE) module to purify cross-task representations and propose a Key Complementary Feature Fusion (KCFF) mechanism to bridge the gap between high-level semantics and low-level visual features, ensuring a balance between task specialization and cross-task enhancement. Extensive experiments demonstrate that LCCC significantly outperforms existing methods in both detection accuracy and description quality, validating the effectiveness and generality of the proposed critique-correction paradigm for synergistic multi-task modeling. The code of the proposed method is available at https://github.com/Throb16/Lccc.
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Image Processing
- Vol. 35, 7783-7798
Learning to Change by Critique and Correction: A Synergistic Framework for Remote Sensing Change Detection and Captioning
Change Detection (CD) and Change Captioning (CC) are two core tasks for understanding land-cover evolution in remote sensing imagery. Existing approaches have explored CD-CC joint modeling through shared representations, task-specific decoding, feature interaction, and semantic guidance. However, the lack of explicit cross-task feedback mechanisms often leads to mutual interference, making it difficult to achieve both accurate detection and expressive descriptions. To address this issue, we propose the Learning to Change by Critique and Correction (LCCC) framework, which reformulates CD and CC as a critique-correction closed-loop process. In LCCC, CD and CC no longer passively share features but interact through bidirectional critique and correction: the CD task provides explicit spatial constraints for CC, and CC, in turn, supervises CD via a Text-Guided Critique Attention (TGCA) mechanism, establishing a synergistic relationship where both tasks act as critics and correctors. Furthermore, we design a Reciprocal Suppression and Enhancement (RSE) module to purify cross-task representations and propose a Key Complementary Feature Fusion (KCFF) mechanism to bridge the gap between high-level semantics and low-level visual features, ensuring a balance between task specialization and cross-task enhancement. Extensive experiments demonstrate that LCCC significantly outperforms existing methods in both detection accuracy and description quality, validating the effectiveness and generality of the proposed critique-correction paradigm for synergistic multi-task modeling. The code of the proposed method is available at https://github.com/Throb16/Lccc.
- 19 November 2025
- Published by Elsevier in Information Fusion
- Vol. 128
Joint multi-view embedding with progressive multi-scale alignment for unaligned infrared-visible image fusion
- 19 November 2025
- Published by Elsevier in Information Fusion
- Vol. 128
Joint multi-view embedding with progressive multi-scale alignment for unaligned infrared-visible image fusion
- 14 November 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Image Processing
- Vol. 34  (10577149) , 7558-7571
Disentangling Inter- and Intra-Video Relations for Multi-Event Video-Text Retrieval and Grounding
Video-text retrieval aims to precisely search for videos most relevant to text queries within a video corpus. However, existing methods are largely limited to single-text (single-event) queries and are not effective at handling multi-text (multi-event) queries. Furthermore, these methods typically focus solely on retrieval and do not attempt to locate multiple events within the retrieved videos. To address these limitations, our paper proposes a novel method named Disentangling Inter- and Intra-Video Relations, which jointly addresses multi-event video-text retrieval and grounding. This method leverages both inter-video and intra-video event relationships to enhance retrieval and grounding performance. At the retrieval level, we devise a Relational Event-Centric Video-Text Retrieval module based on the principle that comprehensive textual information leads to precise correspondence between text and video. It incorporates event relationship features at different hierarchical levels and exploits the hierarchical structure of video relationships to achieve multi-level contrastive learning between events and videos. This approach enhances the richness, accuracy, and comprehensiveness of event descriptions, improving alignment precision between text and video and enabling effective differentiation among videos. For event grounding, we propose Event Contrast-Driven Video Grounding, which accounts for positional differences among events on the 2D temporal score map and achieves precise grounding of multiple events through divergence learning for their locations. Our solution not only provides efficient text-to-video retrieval but also accurately grounds events within the retrieved videos, addressing the shortcomings of existing methods. Extensive experimental results on the ActivityNet Captions and Charades-STA benchmark datasets demonstrate the superior performance of our method, validating its effectiveness. The innovation of this research lies in introducing a new joint framework for video-text retrieval and multi-event grounding while offering new ideas for further research and applications in related fields. The code is available at https://github.com/X7J92/MVT-RG.
- 14 November 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Image Processing
- Vol. 34  (10577149) , 7558-7571
Disentangling Inter- and Intra-Video Relations for Multi-Event Video-Text Retrieval and Grounding
Video-text retrieval aims to precisely search for videos most relevant to text queries within a video corpus. However, existing methods are largely limited to single-text (single-event) queries and are not effective at handling multi-text (multi-event) queries. Furthermore, these methods typically focus solely on retrieval and do not attempt to locate multiple events within the retrieved videos. To address these limitations, our paper proposes a novel method named Disentangling Inter- and Intra-Video Relations, which jointly addresses multi-event video-text retrieval and grounding. This method leverages both inter-video and intra-video event relationships to enhance retrieval and grounding performance. At the retrieval level, we devise a Relational Event-Centric Video-Text Retrieval module based on the principle that comprehensive textual information leads to precise correspondence between text and video. It incorporates event relationship features at different hierarchical levels and exploits the hierarchical structure of video relationships to achieve multi-level contrastive learning between events and videos. This approach enhances the richness, accuracy, and comprehensiveness of event descriptions, improving alignment precision between text and video and enabling effective differentiation among videos. For event grounding, we propose Event Contrast-Driven Video Grounding, which accounts for positional differences among events on the 2D temporal score map and achieves precise grounding of multiple events through divergence learning for their locations. Our solution not only provides efficient text-to-video retrieval but also accurately grounds events within the retrieved videos, addressing the shortcomings of existing methods. Extensive experimental results on the ActivityNet Captions and Charades-STA benchmark datasets demonstrate the superior performance of our method, validating its effectiveness. The innovation of this research lies in introducing a new joint framework for video-text retrieval and multi-event grounding while offering new ideas for further research and applications in related fields. The code is available at https://github.com/X7J92/MVT-RG.
- Open Access
- 27 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 5247-5256
Dual-Granularity Cross-Modal Identity Association for Weakly-Supervised Text-to-Person Image Matching
- 27 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 5247-5256
Open AccessDual-Granularity Cross-Modal Identity Association for Weakly-Supervised Text-to-Person Image Matching