Yajun Qiao
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Research Interests
Yajun Qiao has a total of 13 co-authors. Here are the top co-authors by number of shared publications:
- Lingjuan Miao
School of Automation, Beijing Institute of Technology, Beijing, China
9 shared publications - Zhiqiang Zhou
School of Automation, Beijing Institute of Technology, Beijing, China
8 shared publications - Yuhao Wang
School of Automation, Beijing Institute of Technology, Beijing, China
7 shared publications - Qi Ming
Beijing University of Technology
2 shared publications Lei Zhang
School of Automation, Beijing Institute of Technology, Beijing, China
1 shared publication- Lei Zhang
School of Computer Science, Beijing Institute of Technology, Beijing, China
1 shared publication
Latest Publications
- Open Access
- 9 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 11046-11054
Faster and Better: Reinforced Collaborative Distillation and Self-Learning for Infrared-Visible Image Fusion
- 9 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 11046-11054
Open AccessFaster and Better: Reinforced Collaborative Distillation and Self-Learning for Infrared-Visible Image Fusion
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Neural Networks and Learning Systems
- Vol. 37 (10) , 1-15
Infrared and Visible Image Fusion With Language-Driven Loss and Knowledge Distillation
Infrared and visible image fusion (IVIF) has attracted much attention owing to the highly complementary properties of the two image modalities. Due to the lack of ground-truth fused images, the fusion output of current deep-learning-based methods heavily depends on the loss functions defined mathematically. As it is hard to well mathematically define the fused image without ground truth, the performance of existing fusion methods is limited. In this article, we first propose to use natural language to express the objective of IVIF, which can avoid the explicit mathematical modeling of fusion output in current losses, and make full use of the advantage of language expression to improve the fusion performance. For this purpose, we present a comprehensive language-expressed fusion objective and encode relevant texts into the multimodal embedding space using CLIP. A language-driven fusion model is then constructed in the embedding space by establishing the relationship among the embedded vectors representing the fusion objective and input image modalities. Finally, a language-driven loss is derived to make the actual IVIF aligned with the embedded language-driven fusion model via supervised training. In addition, to ensure that the loss works well with high robustness and generalization in practice, we introduce a novel regularization and patch filtering approach in the training. For real-time applications, we further propose a novel knowledge distillation framework, in which the student conducts self-learning under the supervision of the teacher that has been trained with the language-driven loss. This enables the student to maintain comparable performance and, more importantly, achieve real-time inference even on high-resolution images. Experiments show that our method can obtain much better fusion results than existing techniques. The code is available at https://github.com/wyhlaowang/LDFusion.
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Neural Networks and Learning Systems
- Vol. 37 (10) , 1-15
Infrared and Visible Image Fusion With Language-Driven Loss and Knowledge Distillation
Infrared and visible image fusion (IVIF) has attracted much attention owing to the highly complementary properties of the two image modalities. Due to the lack of ground-truth fused images, the fusion output of current deep-learning-based methods heavily depends on the loss functions defined mathematically. As it is hard to well mathematically define the fused image without ground truth, the performance of existing fusion methods is limited. In this article, we first propose to use natural language to express the objective of IVIF, which can avoid the explicit mathematical modeling of fusion output in current losses, and make full use of the advantage of language expression to improve the fusion performance. For this purpose, we present a comprehensive language-expressed fusion objective and encode relevant texts into the multimodal embedding space using CLIP. A language-driven fusion model is then constructed in the embedding space by establishing the relationship among the embedded vectors representing the fusion objective and input image modalities. Finally, a language-driven loss is derived to make the actual IVIF aligned with the embedded language-driven fusion model via supervised training. In addition, to ensure that the loss works well with high robustness and generalization in practice, we introduce a novel regularization and patch filtering approach in the training. For real-time applications, we further propose a novel knowledge distillation framework, in which the student conducts self-learning under the supervision of the teacher that has been trained with the language-driven loss. This enables the student to maintain comparable performance and, more importantly, achieve real-time inference even on high-resolution images. Experiments show that our method can obtain much better fusion results than existing techniques. The code is available at https://github.com/wyhlaowang/LDFusion.
- Open Access
- 27 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 1443-1451
Infrared and Visible Image Fusion with Language-Driven Loss in CLIP Embedding Space
- 27 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 1443-1451
Open AccessInfrared and Visible Image Fusion with Language-Driven Loss in CLIP Embedding Space
- Open Access
Faster and Better: Reinforced Collaborative Distillation and Self-Learning for Infrared-Visible Image Fusion
Infrared and visible image fusion plays a critical role in enhancing scene perception by combining complementary information from different modalities. Despite recent advances, achieving high-quality image fusion with lightweight models remains a significant challenge. To bridge this gap, we propose a novel collaborative distillation and self-learning framework for image fusion driven by reinforcement learning. Unlike conventional distillation, this approach not only enables the student model to absorb image fusion knowledge from the teacher model, but more importantly, allows the student to perform self-learning on more challenging samples to enhance its capabilities. Particularly, in our framework, a reinforcement learning agent explores and identifies a more suitable training strategy for the student. The agent takes both the student's performance and the teacher-student gap as inputs, which leads to the generation of challenging samples to facilitate the student's self-learning. Simultaneously, it dynamically adjusts the teacher's guidance strength based on the student's state to optimize the knowledge transfer. Experimental results demonstrate that our method can significantly improve student performance and achieve better fusion results compared to existing techniques.
Open AccessFaster and Better: Reinforced Collaborative Distillation and Self-Learning for Infrared-Visible Image Fusion
Infrared and visible image fusion plays a critical role in enhancing scene perception by combining complementary information from different modalities. Despite recent advances, achieving high-quality image fusion with lightweight models remains a significant challenge. To bridge this gap, we propose a novel collaborative distillation and self-learning framework for image fusion driven by reinforcement learning. Unlike conventional distillation, this approach not only enables the student model to absorb image fusion knowledge from the teacher model, but more importantly, allows the student to perform self-learning on more challenging samples to enhance its capabilities. Particularly, in our framework, a reinforcement learning agent explores and identifies a more suitable training strategy for the student. The agent takes both the student's performance and the teacher-student gap as inputs, which leads to the generation of challenging samples to facilitate the student's self-learning. Simultaneously, it dynamically adjusts the teacher's guidance strength based on the student's state to optimize the knowledge transfer. Experimental results demonstrate that our method can significantly improve student performance and achieve better fusion results compared to existing techniques.
- 22 August 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 63 (01962892) , 1-17
Unified Five-Distance Bounding Box Representation for Remote Sensing Oriented Object Detection
Currently, due to the densely packed and varied orientations of objects in the remote sensing images, the oriented object detection is still a challenging task. In the existing methods, the oriented bounding box (OBB) representations either include angle parameters or require a larger number of parameters and additional processes to obtain the final rotated rectangles, which can potentially induce the positional errors and decrease the accuracy of object detection in remote sensing images. In this paper, we propose a unified five-distance bounding box representation for remote sensing oriented object detection, which gets rid of the utilization of angle parameters and simultaneously represents the OBBs with the least distance-based parameters. We then construct a unified five-distance region proposal network (UF-RPN) based on the new OBB representation to achieve higher performance of oriented object detection. Additionally, most existing two-stage methods assume that the localization errors of objects are relatively small in the second stage of the network and believe that directly regressing the errors of general OBB representation would achieve sufficiently-high performance of object detection, neglecting that there is still room for improvement in the detection head by utilizing a more advanced OBB representation. Therefore, a five-distance detection head (FD-Head) is developed by utilizing the five-distance OBB representation in the detection head. Experimental results on the HRSC2016, DIOR-R, DOTA-v1.0 and DOTA-v2.0 datasets demonstrate the superiority and robustness of our method. Our code is publicly available at https://github.com/QYJ123/UFDet.
- 22 August 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 63 (01962892) , 1-17
Unified Five-Distance Bounding Box Representation for Remote Sensing Oriented Object Detection
Currently, due to the densely packed and varied orientations of objects in the remote sensing images, the oriented object detection is still a challenging task. In the existing methods, the oriented bounding box (OBB) representations either include angle parameters or require a larger number of parameters and additional processes to obtain the final rotated rectangles, which can potentially induce the positional errors and decrease the accuracy of object detection in remote sensing images. In this paper, we propose a unified five-distance bounding box representation for remote sensing oriented object detection, which gets rid of the utilization of angle parameters and simultaneously represents the OBBs with the least distance-based parameters. We then construct a unified five-distance region proposal network (UF-RPN) based on the new OBB representation to achieve higher performance of oriented object detection. Additionally, most existing two-stage methods assume that the localization errors of objects are relatively small in the second stage of the network and believe that directly regressing the errors of general OBB representation would achieve sufficiently-high performance of object detection, neglecting that there is still room for improvement in the detection head by utilizing a more advanced OBB representation. Therefore, a five-distance detection head (FD-Head) is developed by utilizing the five-distance OBB representation in the detection head. Experimental results on the HRSC2016, DIOR-R, DOTA-v1.0 and DOTA-v2.0 datasets demonstrate the superiority and robustness of our method. Our code is publicly available at https://github.com/QYJ123/UFDet.
- Open Access
- 10 March 2024
- Published by MDPI AG in Remote Sensing
- Vol. 16 (6) , 969
Infrared–Visible Image Fusion through Feature-Based Decomposition and Domain Normalization
Infrared–visible image fusion is valuable across various applications due to the complementary information that it provides. However, the current fusion methods face challenges in achieving high-quality fused images. This paper identifies a limitation in the existing fusion framework that affects the fusion quality: modal differences between infrared and visible images are often overlooked, resulting in the poor fusion of the two modalities. This limitation implies that features from different sources may not be consistently fused, which can impact the quality of the fusion results. Therefore, we propose a framework that utilizes feature-based decomposition and domain normalization. This decomposition method separates infrared and visible images into common and unique regions. To reduce modal differences while retaining unique information from the source images, we apply domain normalization to the common regions within the unified feature space. This space can transform infrared features into a pseudo-visible domain, ensuring that all features are fused within the same domain and minimizing the impact of modal differences during the fusion process. Noise in the source images adversely affects the fused images, compromising the overall fusion performance. Thus, we propose the non-local Gaussian filter. This filter can learn the shape and parameters of its filtering kernel based on the image features, effectively removing noise while preserving details. Additionally, we propose a novel dense attention in the feature extraction module, enabling the network to understand and leverage inter-layer information. Our experiments demonstrate a marked improvement in fusion quality with our proposed method.
- 10 March 2024
- Published by MDPI AG in Remote Sensing
- Vol. 16 (6) , 969
Open AccessInfrared–Visible Image Fusion through Feature-Based Decomposition and Domain Normalization
Infrared–visible image fusion is valuable across various applications due to the complementary information that it provides. However, the current fusion methods face challenges in achieving high-quality fused images. This paper identifies a limitation in the existing fusion framework that affects the fusion quality: modal differences between infrared and visible images are often overlooked, resulting in the poor fusion of the two modalities. This limitation implies that features from different sources may not be consistently fused, which can impact the quality of the fusion results. Therefore, we propose a framework that utilizes feature-based decomposition and domain normalization. This decomposition method separates infrared and visible images into common and unique regions. To reduce modal differences while retaining unique information from the source images, we apply domain normalization to the common regions within the unified feature space. This space can transform infrared features into a pseudo-visible domain, ensuring that all features are fused within the same domain and minimizing the impact of modal differences during the fusion process. Noise in the source images adversely affects the fused images, compromising the overall fusion performance. Thus, we propose the non-local Gaussian filter. This filter can learn the shape and parameters of its filtering kernel based on the image features, effectively removing noise while preserving details. Additionally, we propose a novel dense attention in the feature extraction module, enabling the network to understand and leverage inter-layer information. Our experiments demonstrate a marked improvement in fusion quality with our proposed method.
- 17 November 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 3200-3205
UFFusion: An Unified Feature Space for Infrared-Visible Image Fusion Network Based on Dynamic Domain Transformation
Infrared-visible images have a high information complementarity, making fusing them highly valuable for various applications. However, infrared-visible images also exhibit strong differences, which are crucial factors limiting fusion performance. To address this issue, we propose a unified feature space in which can transfer the infrared domain to the visible domain using the dynamic domain transformation method. This approach eliminates the modality differences and provides high-quality features for image reconstructor. Notably, we propose a dense attention module used to extract common and unique features. The method permits the model to learn the correlation of different layer features, thereby enhancing the model's performance. Moreover, we design a $\mathbf{S}^{3}\mathbf{IM}$ loss function to enhance dynamic range of fused images. The qualitative and quantitative experiments on publicly available datasets demonstrate the superiority of our UFFusion over the state-of-the-art, in terms of both visual effect and quantitative metrics.- 17 November 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 3200-3205
UFFusion: An Unified Feature Space for Infrared-Visible Image Fusion Network Based on Dynamic Domain Transformation
Infrared-visible images have a high information complementarity, making fusing them highly valuable for various applications. However, infrared-visible images also exhibit strong differences, which are crucial factors limiting fusion performance. To address this issue, we propose a unified feature space in which can transfer the infrared domain to the visible domain using the dynamic domain transformation method. This approach eliminates the modality differences and provides high-quality features for image reconstructor. Notably, we propose a dense attention module used to extract common and unique features. The method permits the model to learn the correlation of different layer features, thereby enhancing the model's performance. Moreover, we design a $\mathbf{S}^{3}\mathbf{IM}$ loss function to enhance dynamic range of fused images. The qualitative and quantitative experiments on publicly available datasets demonstrate the superiority of our UFFusion over the state-of-the-art, in terms of both visual effect and quantitative metrics.- 3 August 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 61 (01962892) , 1-15
A Novel Object Detector Based on High-Quality Rotation Proposal Generation and Adaptive Angle Optimization
Currently, reliable and accurate oriented detection in remote sensing images still needs to be improved. The wide variation of object shapes and orientations in the remote sensing images usually leads to two issues in two-stage oriented object detectors. One issue is how to generate high-quality rotation proposals. The other is the angular error sensitivity to the aspect ratios in the angle optimization process. In this paper, we propose a novel rotation proposal generation and optimization detector, which is based on high-quality rotation proposal generation and adaptive angle optimization to solve these two issues. The proposed method mainly establishes the geometric relationship guided region proposal networks (GRG-RPN) and the adaptive angle optimization head (AAO-Head) to achieve more accurate oriented object detection. The GRG-RPN only uses a simple network and a small number of horizontal anchors to predict high-quality rotation proposals. This approach was derived via the calculation based on the theoretical analysis of the geometric relationship between the oriented bounding boxes (OBBs) and their external horizontal bounding boxes (EHBBs). The AAO-Head solves the angular error sensitivity to the aspect ratios and achieves adaptive angle optimization using a new regression parameter, which is defined based on the theoretical analysis of the relationship between the Intersection over Union (IoU), the angular errors and the aspect ratios. Experiments show that our method can achieve 2.5% mAP improvement averagely versus the compared SOTA methods, and achieve 0.5% mAP improvement versus the next best method Oriented R-CNN with fewer regression parameters and a simpler regression approach.
- 3 August 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 61 (01962892) , 1-15
A Novel Object Detector Based on High-Quality Rotation Proposal Generation and Adaptive Angle Optimization
Currently, reliable and accurate oriented detection in remote sensing images still needs to be improved. The wide variation of object shapes and orientations in the remote sensing images usually leads to two issues in two-stage oriented object detectors. One issue is how to generate high-quality rotation proposals. The other is the angular error sensitivity to the aspect ratios in the angle optimization process. In this paper, we propose a novel rotation proposal generation and optimization detector, which is based on high-quality rotation proposal generation and adaptive angle optimization to solve these two issues. The proposed method mainly establishes the geometric relationship guided region proposal networks (GRG-RPN) and the adaptive angle optimization head (AAO-Head) to achieve more accurate oriented object detection. The GRG-RPN only uses a simple network and a small number of horizontal anchors to predict high-quality rotation proposals. This approach was derived via the calculation based on the theoretical analysis of the geometric relationship between the oriented bounding boxes (OBBs) and their external horizontal bounding boxes (EHBBs). The AAO-Head solves the angular error sensitivity to the aspect ratios and achieves adaptive angle optimization using a new regression parameter, which is defined based on the theoretical analysis of the relationship between the Intersection over Union (IoU), the angular errors and the aspect ratios. Experiments show that our method can achieve 2.5% mAP improvement averagely versus the compared SOTA methods, and achieve 0.5% mAP improvement versus the next best method Oriented R-CNN with fewer regression parameters and a simpler regression approach.