Songwei Pei
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Songwei Pei has a total of 47 co-authors. Here are the top co-authors by number of shared publications:
- Shangguang Wang
Beijing University of Posts and Telecommunications
11 shared publications - Qian Li
Beijing University of Posts and Telecommunications, Beijing, China
8 shared publications - Bingfeng Liu
School of Computer Science, Beijing University of Posts and Telecommunications, Beijing, China
7 shared publications - Yuanzhou Huang
Beijing University of Posts and Telecommunications
5 shared publications - Shuhuai Wang
School of Computer Science, Beijing University of Posts and Telecommunications, Beijing, China
4 shared publications Shuhuai Wang
3 shared publications
Latest Publications
- Open Access
- 9 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 6118-6127
CLIPFuse: Transfer CLIP to General Image Fusion via Task Awareness
- 9 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 6118-6127
Open AccessCLIPFuse: Transfer CLIP to General Image Fusion via Task Awareness
- 4 August 2026
- Published by Elsevier in Computer Vision and Image Understanding
- Vol. 271
DepthFusion: Depth-assisted infrared–visible image fusion
- 4 August 2026
- Published by Elsevier in Computer Vision and Image Understanding
- Vol. 271
DepthFusion: Depth-assisted infrared–visible image fusion
- Open Access
- 26 July 2026
- Published by Association for Computing Machinery (ACM)
- p. 1-7
ATLAS: Asynchronous Topological Learning for Accurate FIT Prediction via Decoupled Graph Neural Networks
- 26 July 2026
- Published by Association for Computing Machinery (ACM)
- p. 1-7
Open AccessATLAS: Asynchronous Topological Learning for Accurate FIT Prediction via Decoupled Graph Neural Networks
- 10 June 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Multimedia
- Vol. PP (15209210) , 1-13
Composite Degradation Image Restoration Based on Multi-Dimensional Continuous Degradation Intensity Prediction
The restoration of degraded images has long been a critical challenge in image preprocessing, as degradation patterns manifest in diverse forms. These include both equipment-related factors, such as sensor noise and motion blur, and environmental factors, such as haze, rain, and low-light conditions. The primary objective of image restoration is to recover pristine visual content from these degraded inputs. In real-world scenarios, multiple types of degradation often coexist in composite forms. However, conventional restoration methods typically focus on single-degradation scenarios, leading to suboptimal performance when addressing complex real-world cases. Furthermore, existing algorithms face significant challenges due to the scarcity of comprehensive datasets that capture composite degradations. To address these limitations, we propose a network named CDG-IR, which uses composite prompt guidance to achieve image restoration under continuous composite degradation factors. The CDG-IR architecture comprises two synergistic components: a Composite Prompt Generator (CPG) that analyzes degradation characteristics and produces a prompt vector, and an Image Restoration Transformer (IRT) that employs customized attention mechanisms to generate restored images. The CPG module generates a composite prompt by identifying intertwined degradation features, while the IRT effectively integrates these prompts with visual features through innovative self-attention and cross-attention modules. To further fill a crucial gap in current data resources for multi-degradation research, we construct MD-83W, a novel composite degradation dataset comprising 830,000 samples with continuous variations across five degradation dimensions. Experimental evaluations on MD-83W and public benchmarks demonstrate the state-of-the-art performance of our method.
- 10 June 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Multimedia
- Vol. PP (15209210) , 1-13
Composite Degradation Image Restoration Based on Multi-Dimensional Continuous Degradation Intensity Prediction
The restoration of degraded images has long been a critical challenge in image preprocessing, as degradation patterns manifest in diverse forms. These include both equipment-related factors, such as sensor noise and motion blur, and environmental factors, such as haze, rain, and low-light conditions. The primary objective of image restoration is to recover pristine visual content from these degraded inputs. In real-world scenarios, multiple types of degradation often coexist in composite forms. However, conventional restoration methods typically focus on single-degradation scenarios, leading to suboptimal performance when addressing complex real-world cases. Furthermore, existing algorithms face significant challenges due to the scarcity of comprehensive datasets that capture composite degradations. To address these limitations, we propose a network named CDG-IR, which uses composite prompt guidance to achieve image restoration under continuous composite degradation factors. The CDG-IR architecture comprises two synergistic components: a Composite Prompt Generator (CPG) that analyzes degradation characteristics and produces a prompt vector, and an Image Restoration Transformer (IRT) that employs customized attention mechanisms to generate restored images. The CPG module generates a composite prompt by identifying intertwined degradation features, while the IRT effectively integrates these prompts with visual features through innovative self-attention and cross-attention modules. To further fill a crucial gap in current data resources for multi-degradation research, we construct MD-83W, a novel composite degradation dataset comprising 830,000 samples with continuous variations across five degradation dimensions. Experimental evaluations on MD-83W and public benchmarks demonstrate the state-of-the-art performance of our method.
- Open Access
- 8 June 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Multimedia Computing, Communications, and Applications
- Vol. 22 (6) , 1-23
MPGNet: Multi-Prompt Guided Diffusion Network for All-in-One Image Restoration
Restoring degraded images has long been a significant focus in image preprocessing, aiming to recover high-quality images from their degraded counterparts. While numerous studies have addressed specific types of degradation, All-in-One methods have achieved a notable milestone by enabling a single model to handle various degradations without requiring prior knowledge of the degradation types. However, these approaches often fall short of delivering satisfactory performance in complex and multiple degradation scenarios. To address these challenges, this article proposes a Multi-Prompt Guided Diffusion Network for All-in-One image restoration, referred to as MPGNet. MPGNet consists of two key components: the Image Restoration Part (IRP) and the Prompt Generation Part (PGP). The IRP employs a diffusion model to perform the image restoration task, while the PGP generates tailored prompts to guide image restoration from degradation. By leveraging meticulously designed prompts, MPGNet empowers the diffusion network to adaptively handle complex and multiple degradation scenarios. To enhance adaptability to various degradation types, content specificity, and overall performance, we design three types of prompts: Style Prompt, Content Prompt, and Learnable Prompt. The first two prompts are carefully generated using the Image Encoder from the pre-trained vision-text model CLIP, with contrastive learning applied during respective training, while the Learnable Prompt is custom-designed separately. We also introduce a Prompt Encoder that integrates Style, Content, and Learnable Prompts into a unified representation, facilitating image restoration for IRP. The experimental results demonstrate that the proposed method significantly outperforms recent approaches in both single and multiple degradation scenarios.
- 8 June 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Multimedia Computing, Communications, and Applications
- Vol. 22 (6) , 1-23
Open AccessMPGNet: Multi-Prompt Guided Diffusion Network for All-in-One Image Restoration
Restoring degraded images has long been a significant focus in image preprocessing, aiming to recover high-quality images from their degraded counterparts. While numerous studies have addressed specific types of degradation, All-in-One methods have achieved a notable milestone by enabling a single model to handle various degradations without requiring prior knowledge of the degradation types. However, these approaches often fall short of delivering satisfactory performance in complex and multiple degradation scenarios. To address these challenges, this article proposes a Multi-Prompt Guided Diffusion Network for All-in-One image restoration, referred to as MPGNet. MPGNet consists of two key components: the Image Restoration Part (IRP) and the Prompt Generation Part (PGP). The IRP employs a diffusion model to perform the image restoration task, while the PGP generates tailored prompts to guide image restoration from degradation. By leveraging meticulously designed prompts, MPGNet empowers the diffusion network to adaptively handle complex and multiple degradation scenarios. To enhance adaptability to various degradation types, content specificity, and overall performance, we design three types of prompts: Style Prompt, Content Prompt, and Learnable Prompt. The first two prompts are carefully generated using the Image Encoder from the pre-trained vision-text model CLIP, with contrastive learning applied during respective training, while the Learnable Prompt is custom-designed separately. We also introduce a Prompt Encoder that integrates Style, Content, and Learnable Prompts into a unified representation, facilitating image restoration for IRP. The experimental results demonstrate that the proposed method significantly outperforms recent approaches in both single and multiple degradation scenarios.
- 17 February 2026
- Published by Elsevier in Computer Vision and Image Understanding
- Vol. 265
SRDR: Style recovery and detail replenishment matter for single image dehazing
- 17 February 2026
- Published by Elsevier in Computer Vision and Image Understanding
- Vol. 265
SRDR: Style recovery and detail replenishment matter for single image dehazing
- Top Cited Papers
- 28 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 64 (01962892) , 1-15
DENet: Dual-Path Edge Network With Global-Local Attention for Infrared Small Target Detection
Infrared small target detection (IRSTD) is crucial for remote sensing applications like disaster warning and maritime surveillance. However, due to the lack of distinctive texture and morphological features, infrared small targets are highly susceptible to blending into cluttered and noisy backgrounds. Existing methods often rely on fixed gradient operators (e.g., Sobel, Canny) or simplistic attention mechanisms, which are inadequate for accurately extracting target edges under low contrast and high noise. In this paper, we propose an enhanced dual-path edge network (DENet) that explicitly addresses this challenge by decoupling edge enhancement and semantic modeling into two deliberately designed processing paths. The first path employs a Bidirectional-Interaction Module (BIM), which uses both Local Self-Attention and Global Self-Attention to capture multi-scale local and global feature dependencies. The global attention mechanism, based on a Transformer architecture, integrates long-range semantic relationships and contextual information, ensuring robust scene understanding. The second path introduces the Multi-Edge Refiner (Multi-ER), which enhances fine-grained edge details through multi-scale cascaded refinement. Coupled with attention-driven gating, it improves edge localization for targets of varying sizes and suppresses noise effectively. Extensive experiments on the IRSTD-1K, NUDT-SIRST and NUAA-SIRST benchmarks demonstrate that DENet significantly outperforms state-of-the-art methods, achieving superior Mean Intersection over Union and pixel-level accuracy, while maintaining lower false alarm rates. The proposed framework effectively improves infrared small target detection and localization through joint semantic modeling and edge refinement.
- 28 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 64 (01962892) , 1-15
Top Cited PapersDENet: Dual-Path Edge Network With Global-Local Attention for Infrared Small Target Detection
Infrared small target detection (IRSTD) is crucial for remote sensing applications like disaster warning and maritime surveillance. However, due to the lack of distinctive texture and morphological features, infrared small targets are highly susceptible to blending into cluttered and noisy backgrounds. Existing methods often rely on fixed gradient operators (e.g., Sobel, Canny) or simplistic attention mechanisms, which are inadequate for accurately extracting target edges under low contrast and high noise. In this paper, we propose an enhanced dual-path edge network (DENet) that explicitly addresses this challenge by decoupling edge enhancement and semantic modeling into two deliberately designed processing paths. The first path employs a Bidirectional-Interaction Module (BIM), which uses both Local Self-Attention and Global Self-Attention to capture multi-scale local and global feature dependencies. The global attention mechanism, based on a Transformer architecture, integrates long-range semantic relationships and contextual information, ensuring robust scene understanding. The second path introduces the Multi-Edge Refiner (Multi-ER), which enhances fine-grained edge details through multi-scale cascaded refinement. Coupled with attention-driven gating, it improves edge localization for targets of varying sizes and suppresses noise effectively. Extensive experiments on the IRSTD-1K, NUDT-SIRST and NUAA-SIRST benchmarks demonstrate that DENet significantly outperforms state-of-the-art methods, achieving superior Mean Intersection over Union and pixel-level accuracy, while maintaining lower false alarm rates. The proposed framework effectively improves infrared small target detection and localization through joint semantic modeling and edge refinement.