Yunqi Wang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Yunqi Wang has a total of 35 co-authors. Here are the top co-authors by number of shared publications:
- Tianjiao Zeng
School of Aeronautics and Astronautics, University of Electronic Science and Technology of China, Chengdu, China
3 shared publications - Yik-Chung Wu
Department of Electrical and Electronic Engineering, The University of Hong Kong, Pokfulam, Hong Kong
3 shared publications - Yang Li
School of Computing and Information Technology, Great Bay University, Dongguan, China
2 shared publications - Bo Ai
Beijing Jiaotong University
1 shared publication - Chengyan Wang
Fudan University
1 shared publication - Chenyang Yang
Beihang University, Beijing, China
1 shared publication
Latest Publications
- 10 January 2026
- Published by Elsevier in Medical Image Analysis
- Vol. 110, 103936
Knowledge-guided multi-geometric window transformer for cardiac cine MRI reconstruction
- 10 January 2026
- Published by Elsevier in Medical Image Analysis
- Vol. 110, 103936
Knowledge-guided multi-geometric window transformer for cardiac cine MRI reconstruction
- Open Access
Large-field-of-view lensless imaging with miniaturized sensors
Lensless cameras replace bulky optics with thin modulation masks, enabling compact imaging systems. However, existing methods rely on an idealized model that assumes a globally shift-invariant point spread function (PSF) and sufficiently large sensors. In reality, the PSF varies spatially across the field of view (FOV), and finite sensor boundaries truncate modulated light--effects that intensify as sensors shrink, degrading peripheral reconstruction quality and limiting the effective FOV. We address these limitations through a local-to-global hierarchical framework grounded in a locally shift-invariant convolution model that explicitly accounts for PSF variation and sensor truncation. Patch-wise learned deconvolution first adaptively estimates local PSFs and reconstructs regions independently. A hierarchical enhancement network then progressively expands its receptive field--from small patches through intermediate blocks to the full image--integrating fine local details with global contextual information. Experiments on public datasets show that our method achieves superior reconstruction quality over a larger effective FOV with significantly reduced sensor sizes. Under extreme miniaturization--sensors reduced to 8% of the original area--we achieve improvements of 2 dB (PSNR) and 5% (SSIM), with particularly notable gains in structural fidelity. Code is available at https://github.com/KB504-public/l2g_lensless_imaging .
Open AccessLarge-field-of-view lensless imaging with miniaturized sensors
Lensless cameras replace bulky optics with thin modulation masks, enabling compact imaging systems. However, existing methods rely on an idealized model that assumes a globally shift-invariant point spread function (PSF) and sufficiently large sensors. In reality, the PSF varies spatially across the field of view (FOV), and finite sensor boundaries truncate modulated light--effects that intensify as sensors shrink, degrading peripheral reconstruction quality and limiting the effective FOV. We address these limitations through a local-to-global hierarchical framework grounded in a locally shift-invariant convolution model that explicitly accounts for PSF variation and sensor truncation. Patch-wise learned deconvolution first adaptively estimates local PSFs and reconstructs regions independently. A hierarchical enhancement network then progressively expands its receptive field--from small patches through intermediate blocks to the full image--integrating fine local details with global contextual information. Experiments on public datasets show that our method achieves superior reconstruction quality over a larger effective FOV with significantly reduced sensor sizes. Under extreme miniaturization--sensors reduced to 8% of the original area--we achieve improvements of 2 dB (PSNR) and 5% (SSIM), with particularly notable gains in structural fidelity. Code is available at https://github.com/KB504-public/l2g_lensless_imaging .
- 30 October 2025
- Published by Elsevier in ISPRS Journal of Photogrammetry and Remote Sensing
- Vol. 231, 151-166
MHFNet: Multimodal hybrid fusion framework for misaligned SAR-Optical ship detection
- 30 October 2025
- Published by Elsevier in ISPRS Journal of Photogrammetry and Remote Sensing
- Vol. 231, 151-166
MHFNet: Multimodal hybrid fusion framework for misaligned SAR-Optical ship detection
- 16 January 2025
- Published by Elsevier in Medical Image Analysis
- Vol. 101, 103459
Illuminating the unseen: Advancing MRI domain generalization through causality
- 16 January 2025
- Published by Elsevier in Medical Image Analysis
- Vol. 101, 103459
Illuminating the unseen: Advancing MRI domain generalization through causality
- 25 October 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Wireless Communications
- Vol. 23 (6) , 5330-5344
ENGNN: A General Edge-Update Empowered GNN Architecture for Radio Resource Management in Wireless Networks
In order to achieve high data rate and ubiquitous connectivity in future wireless networks, a key task is to efficiently manage the radio resource by judicious beamforming and power allocation. Unfortunately, the iterative nature of the commonly applied optimization-based algorithms cannot meet the low latency requirements due to the high computational complexity. For real-time implementations, deep learning-based approaches, especially the graph neural networks (GNNs), have been demonstrated with good scalability and generalization performance due to the permutation equivariance (PE) property. However, the current architectures are only equipped with the node-update mechanism, which prohibits the applications to a more general setup, where the unknown variables are also defined on the graph edges. To fill this gap, we propose an edge-update mechanism, which enables GNNs to handle both node and edge variables and prove its PE property with respect to both transmitters and receivers. Simulation results on typical radio resource management problems demonstrate that the proposed method achieves higher sum rate but with much shorter computation time than state-of-the-art methods and generalizes well on different numbers of base stations and users, different noise variances, interference levels, and transmit power budgets.
- 25 October 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Wireless Communications
- Vol. 23 (6) , 5330-5344
ENGNN: A General Edge-Update Empowered GNN Architecture for Radio Resource Management in Wireless Networks
In order to achieve high data rate and ubiquitous connectivity in future wireless networks, a key task is to efficiently manage the radio resource by judicious beamforming and power allocation. Unfortunately, the iterative nature of the commonly applied optimization-based algorithms cannot meet the low latency requirements due to the high computational complexity. For real-time implementations, deep learning-based approaches, especially the graph neural networks (GNNs), have been demonstrated with good scalability and generalization performance due to the permutation equivariance (PE) property. However, the current architectures are only equipped with the node-update mechanism, which prohibits the applications to a more general setup, where the unknown variables are also defined on the graph edges. To fill this gap, we propose an edge-update mechanism, which enables GNNs to handle both node and edge variables and prove its PE property with respect to both transmitters and receivers. Simulation results on typical radio resource management problems demonstrate that the proposed method achieves higher sum rate but with much shorter computation time than state-of-the-art methods and generalizes well on different numbers of base stations and users, different noise variances, interference levels, and transmit power budgets.
- 12 December 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Image Processing
- Vol. 32 (10577149) , 235-250
Contrastive-ACE: Domain Generalization Through Alignment of Causal Mechanisms
Domain generalization aims to learn knowledge invariant across different distributions while semantically meaningful for downstream tasks from multiple source domains, to improve the model’s generalization ability on unseen target domains. The fundamental objective is to understand the underlying ”invariance” behind these observational distributions and such invariance has been shown to have a close connection to causality. While many existing approaches make use of the property that causal features are invariant across domains, we consider the invariance of the average causal effect of the features to the labels. This invariance regularizes our training approach in which interventions are performed on features to enforce stability of the causal prediction by the classifier across domains. Our work thus sheds some light on the domain generalization problem by introducing invariance of the mechanisms into the learning process. Experiments on several benchmark datasets demonstrate the performance of the proposed method against SOTAs. The codes are available at: https://github.com/lithostark/Contrastive-ACE .
- 12 December 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Image Processing
- Vol. 32 (10577149) , 235-250
Contrastive-ACE: Domain Generalization Through Alignment of Causal Mechanisms
Domain generalization aims to learn knowledge invariant across different distributions while semantically meaningful for downstream tasks from multiple source domains, to improve the model’s generalization ability on unseen target domains. The fundamental objective is to understand the underlying ”invariance” behind these observational distributions and such invariance has been shown to have a close connection to causality. While many existing approaches make use of the property that causal features are invariant across domains, we consider the invariance of the average causal effect of the features to the labels. This invariance regularizes our training approach in which interventions are performed on features to enforce stability of the causal prediction by the classifier across domains. Our work thus sheds some light on the domain generalization problem by introducing invariance of the mechanisms into the learning process. Experiments on several benchmark datasets demonstrate the performance of the proposed method against SOTAs. The codes are available at: https://github.com/lithostark/Contrastive-ACE .
- 8 November 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Wireless Communications
- Vol. 22 (5) , 3432-3446
Heterogeneous Transformer: A Scale Adaptable Neural Network Architecture for Device Activity Detection
To support modern machine-type communications, a crucial task during the random access phase is device activity detection, which is to identify the active devices from a large number of potential devices based on the received signal at the access point. By utilizing the statistical properties of the channel, state-of-the-art covariance based methods have been demonstrated to achieve better activity detection performance than compressed sensing based methods. However, covariance based methods require to solve a high dimensional nonconvex optimization problem by updating the estimate of the activity status of each device sequentially. Since the number of updates is proportional to the device number, the computational complexity and delay make the iterative updates difficult for real-time implementation especially when the device number scales up. Inspired by the success of deep learning for real-time inference, this paper proposes a learning based method with a customized heterogeneous transformer architecture for device activity detection. By adopting an attention mechanism in the architecture design, the proposed method is able to extract features reflecting relevance among device pilots and received signal, permutation equivariant with respect to devices, and its training parameter number is independent of the device number. Simulation results demonstrate that the proposed method achieves better activity detection performance with much shorter computation time than state-of-the-art covariance approach, and generalizes well to different numbers of devices and BS-antennas, different pilot lengths, transmit powers, and cell radii.
- 8 November 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Wireless Communications
- Vol. 22 (5) , 3432-3446
Heterogeneous Transformer: A Scale Adaptable Neural Network Architecture for Device Activity Detection
To support modern machine-type communications, a crucial task during the random access phase is device activity detection, which is to identify the active devices from a large number of potential devices based on the received signal at the access point. By utilizing the statistical properties of the channel, state-of-the-art covariance based methods have been demonstrated to achieve better activity detection performance than compressed sensing based methods. However, covariance based methods require to solve a high dimensional nonconvex optimization problem by updating the estimate of the activity status of each device sequentially. Since the number of updates is proportional to the device number, the computational complexity and delay make the iterative updates difficult for real-time implementation especially when the device number scales up. Inspired by the success of deep learning for real-time inference, this paper proposes a learning based method with a customized heterogeneous transformer architecture for device activity detection. By adopting an attention mechanism in the architecture design, the proposed method is able to extract features reflecting relevance among device pilots and received signal, permutation equivariant with respect to devices, and its training parameter number is independent of the device number. Simulation results demonstrate that the proposed method achieves better activity detection performance with much shorter computation time than state-of-the-art covariance approach, and generalizes well to different numbers of devices and BS-antennas, different pilot lengths, transmit powers, and cell radii.
- 28 February 2013
- Published by AICIT in INTERNATIONAL JOURNAL ON Advances in Information Sciences and Service Sciences
- Vol. 5 (4) , 246-253
Interaction Design Research on Information Management System on Mobile Phone for Rural Users
- 28 February 2013
- Published by AICIT in INTERNATIONAL JOURNAL ON Advances in Information Sciences and Service Sciences
- Vol. 5 (4) , 246-253
Interaction Design Research on Information Management System on Mobile Phone for Rural Users