Congwei Zhang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Congwei Zhang has a total of 8 co-authors. Here are the top co-authors by number of shared publications:
- Wankou Yang
Southeast University
4 shared publications - Zhibin Quan
School of Automation, Southeast University, Nanjing 210096, China
2 shared publications - Chenning Ma
Department of Acoustics and Microwave Physics, College of Physics, Jilin University, Changchun, Jilin 130012, China
1 shared publication Jiren Mai
Southeast University,School of Automation,Nanjing,China
1 shared publication- Liying Zhang
MIIT Key Laboratory of Pattern Analysis and Machine Intelligence, College of Artificial Intelligence, Nanjing University of Aeronautics and Astronautics, Nanjing, 211106, China
1 shared publication - Xian Zhang
Southeast Unversity, Nanjing, China
1 shared publication
Latest Publications
- 8 October 2026
- Published by Elsevier in Expert Systems with Applications
- Vol. 334
Beyond post-hoc alignment: Text-conditioned state refinement for weakly supervised referring image segmentation
- 8 October 2026
- Published by Elsevier in Expert Systems with Applications
- Vol. 334
Beyond post-hoc alignment: Text-conditioned state refinement for weakly supervised referring image segmentation
- 18 March 2026
- Published by Elsevier in Expert Systems with Applications
- Vol. 319
Cross-modal semantic token alignment via contrastive learning for weakly-supervised referring image segmentation
- 18 March 2026
- Published by Elsevier in Expert Systems with Applications
- Vol. 319
Cross-modal semantic token alignment via contrastive learning for weakly-supervised referring image segmentation
- Open Access
- 11 October 2024
- Published by Association for Computing Machinery (ACM)
- p. 88-94
Weakly Supervised Semantic Segmentation via Mamba Image Restoration
Weakly Supervised Semantic Segmentation (WSSS) aims to achieve pixel-level segmentation using easily annotated weak labels to reduce annotation costs. However, existing WSSS methods typically rely on CAM (Class Activation Maps) as an intermediate segmentation step. The concentrated activation regions of CAM limit the final segmentation performance. To address this issue, we treat the CAMs obtained through traditional methods as low-quality images to be restored and design an image restoration method based on the mamba model. This method expands the activation regions guided by the multi-stage CAMs we extract. Extensive experiments conducted on the PASCAL VOC 2012 and MS COCO 2014 segmentation datasets demonstrate that our approach effectively improves the quality of CAMs and significantly enhances the final segmentation performance. Our segmentation performance is very close to the recent state-of-the-art, which proves the effectiveness of our method.
- 11 October 2024
- Published by Association for Computing Machinery (ACM)
- p. 88-94
Open AccessWeakly Supervised Semantic Segmentation via Mamba Image Restoration
Weakly Supervised Semantic Segmentation (WSSS) aims to achieve pixel-level segmentation using easily annotated weak labels to reduce annotation costs. However, existing WSSS methods typically rely on CAM (Class Activation Maps) as an intermediate segmentation step. The concentrated activation regions of CAM limit the final segmentation performance. To address this issue, we treat the CAMs obtained through traditional methods as low-quality images to be restored and design an image restoration method based on the mamba model. This method expands the activation regions guided by the multi-stage CAMs we extract. Extensive experiments conducted on the PASCAL VOC 2012 and MS COCO 2014 segmentation datasets demonstrate that our approach effectively improves the quality of CAMs and significantly enhances the final segmentation performance. Our segmentation performance is very close to the recent state-of-the-art, which proves the effectiveness of our method.
- 16 September 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 24 (21) , 35068-35080
Hierarchical Activation Dual-Backbone Network for Weakly Supervised Semantic Segmentation
Weakly Supervised Semantic Segmentation (WSSS) aims to achieve segmentation using weak labels, thereby reducing annotation costs. Current mainstream WSSS methods utilize Class Activation Map (CAM) to generate pseudo-labels. However, CAM primarily focuses on discriminative regions beneficial for classification, which adversely affects the final segmentation performance. In this paper, inspired by learning mechanism in cognitive science theories, we propose a novel Hierarchical Activation Dual Backbone Network (HADBN) to obtain high-quality CAM. HADBN employs serial dual-backbones to decompose the CAM activation process into a free-training main-body learning part and an iterative detail supply part. Through this decomposition, HADBN highlights detailed regions that conventional CAM often overlooks, and achieves significant performance improvements through the precise application of algorithms. Furthermore, HADBN only takes supervised training with lightweight module on the main-body part of CAM to avoid introducing excessive computational cost. In extension experiments on the VOC and COCO datasets, HADBN significantly improves the quality of CAM and achieves final segmentation results with mIoU scores of 73.6 and 45.3, respectively, which are competitive with recent state-of-the-art methods. Additionally, the framework of HADBN demonstrates a certain level of versatility, making it well-suited for adaptation to other methods and improving their performance.
- 16 September 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 24 (21) , 35068-35080
Hierarchical Activation Dual-Backbone Network for Weakly Supervised Semantic Segmentation
Weakly Supervised Semantic Segmentation (WSSS) aims to achieve segmentation using weak labels, thereby reducing annotation costs. Current mainstream WSSS methods utilize Class Activation Map (CAM) to generate pseudo-labels. However, CAM primarily focuses on discriminative regions beneficial for classification, which adversely affects the final segmentation performance. In this paper, inspired by learning mechanism in cognitive science theories, we propose a novel Hierarchical Activation Dual Backbone Network (HADBN) to obtain high-quality CAM. HADBN employs serial dual-backbones to decompose the CAM activation process into a free-training main-body learning part and an iterative detail supply part. Through this decomposition, HADBN highlights detailed regions that conventional CAM often overlooks, and achieves significant performance improvements through the precise application of algorithms. Furthermore, HADBN only takes supervised training with lightweight module on the main-body part of CAM to avoid introducing excessive computational cost. In extension experiments on the VOC and COCO datasets, HADBN significantly improves the quality of CAM and achieves final segmentation results with mIoU scores of 73.6 and 45.3, respectively, which are competitive with recent state-of-the-art methods. Additionally, the framework of HADBN demonstrates a certain level of versatility, making it well-suited for adaptation to other methods and improving their performance.
- 27 August 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 34, 653-658
ECT: Enhancement Cross Training for Weakly Supervised Semantic Segmentation
In recent years, Weakly Supervised Semantic Segmentation (WSSS) has garnered significant attention, as it enables pixel-level segmentation using only image-level labels. However, current WSSS methods typically rely on extracting Class Activation Map (CAM) from a classification network as the initial localization cues, which are often narrow and fragmented. In this paper, we demonstrate that more regions can be activated by roughly applying enhancement functions on CAM. Specifically, we propose an Enhancement Cross Training (ECT) approach for WSSS, which involves non-learning enhancement functions and a Cross Training process for integrating enhancement functions into the learnable CAM generation network. By cross-training two identical CAM generation models, ECT allows CAM to expand with its own localization information. Experiment on PASCAL VOC 2012 shows that our method is competitive with existing state-of-the-art methods.
- 27 August 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 34, 653-658
ECT: Enhancement Cross Training for Weakly Supervised Semantic Segmentation
In recent years, Weakly Supervised Semantic Segmentation (WSSS) has garnered significant attention, as it enables pixel-level segmentation using only image-level labels. However, current WSSS methods typically rely on extracting Class Activation Map (CAM) from a classification network as the initial localization cues, which are often narrow and fragmented. In this paper, we demonstrate that more regions can be activated by roughly applying enhancement functions on CAM. Specifically, we propose an Enhancement Cross Training (ECT) approach for WSSS, which involves non-learning enhancement functions and a Cross Training process for integrating enhancement functions into the learnable CAM generation network. By cross-training two identical CAM generation models, ECT allows CAM to expand with its own localization information. Experiment on PASCAL VOC 2012 shows that our method is competitive with existing state-of-the-art methods.