I received my Ph.D. degree at the School of Information and Electronics & School of Computer Science and Technology, Beijing Institute of Technology (BIT), supervised by Prof. Ying Fu. I am also fortunate to collaborate closely with Hong-yu Zhou and Lin Gu. I received my M.S. degree from BIT in 2021 and my B.S. degree from China University of Geosciences (Beijing) in 2019.
My research interests primarily lie in computer vision and deep learning, including:
- Frequency Learning
- Image Segmentation, Object Detection
- Low-light Image Enhancement and Recognition
- Remote Sensing Image Processing
- Representation Learning for Transformers & CNNs
- Medical Large Language Models
You can find my full publication list on my Google Scholar profile.
I am open for collaborations in research, especially in the fields of Image Understanding, Representation Learning and Medical LLMs. Please don’t hesitate to get in touch if my research interests you.
I am actively seeking positions in both academia and industry, starting in the Fall of 2025.
🔥 News
-
[2025.10.18] The core concept of our FreqFusion was adopted by WeaveSeg for nuclei instance segmentation, accepted by the International Conference on Computer Vision (ICCV) 2025. [Paper]
-
[2025.9.02] FreqFusion was utilized as a core component for feature aggregation in a new diffusion-based framework for retinal image segmentation, accepted by IEEE Transactions on Medical Imaging (TMI). [Paper]
-
[2025.06.18] Our method, FreqFusion, was successfully applied in ModFusion to enhance esophageal cancer segmentation, published in the journal Knowledge-Based Systems (KBS). [Paper]
-
[2025.04.09] Our method, FreqFusion, was adopted by ACVLab in their solution for the NTIRE 2025 Reflection Removal Challenge (CVPR 2025 Workshop), which secured a top 2% global ranking (6th/244) 🏆. [Event] [Tech Report]
🎓 Education
- 2021.09 - 2025.06,
Beijing Institute of Technology, Ph.D. in Electronic Information - 2019.09 - 2021.06, Beijing Institute of Technology, M.S. in Software Engineering
- 2014.09 - 2019.06,
China University of Geosciences (Beijing), B.S. in Mechanical Engineering
📝 Selected Publications
Frequency-Dynamic Attention Modulation for Dense Prediction
Linwei Chen, Lin Gu, Ying Fu*.
International Conference on Computer Vision (ICCV), 2025.
[PDF] 
🧐 Revealing a “frequency vanishing” phenomenon in ViTs and introducing a frequency-dynamic attention mechanism that strengthens feature representations for dense prediction.
Spatial Frequency Modulation for Semantic Segmentation
Linwei Chen, Ying Fu*, Lin Gu, Dezhi Zheng, Jifeng Dai.
IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 2025.
[PDF] [IEEE] [公众号] 
😎 Proposing Spatial Frequency Modulation (SFM) to mitigate the degradation of high-frequency representations during widely-used downsampling.
Frequency Dynamic Convolution for Dense Image Prediction
Linwei Chen, Lin Gu, Liang Li, Chenggang Yan, Ying Fu*.
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2025.
[PDF] [Zhihu 知乎] 
🤭 Presenting a parameter-efficient frequency dynamic convolution (FDConv) operating in the frequency domain, enabling 64× kernels without increasing parameter cost.
Frequency-aware Feature Fusion for Dense Image Prediction
Linwei Chen, Ying Fu*, Lin Gu, Chenggang Yan, Tatsuya Harada, Gao Huang.
IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 2024.
(WOS Citation Top 0.3%, ICIC2025 Invited Hilight Paper)
[PDF] [Zhihu 知乎] 
🥳 Tackling feature fusion by analyzing frequency characteristics, proposing a plug&play method FrequFusion to effectively combine features for dense prediction.
Frequency-Adaptive Dilated Convolution for Semantic Segmentation
Linwei Chen, Lin Gu, Ying Fu*.
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2024.
(Highlight Poster, Acceptance Rate 2.8%, WOS Citation Top 0.2%)
[PDF] [Zhihu 知乎] 
🤓 Introducing Frequency Adaptive Dilated Convolution (FADC) that adaptively sets its dilation rate from frequency cues, intelligently expanding the receptive field while preserving high-frequency details.
When Semantic Segmentation Meets Frequency Aliasing
Linwei Chen, Lin Gu, Ying Fu*.
International Conference on Learning Representations (ICLR), 2024.
[PDF] [Zhihu 知乎] 
😆 Providing the first systematic study of the frequency aliasing and hard pixel problem in semantic segmentation and proposes a simple yet effective solution.
Instance Segmentation in the Dark
Linwei Chen, Ying Fu*, Kaixuan Wei, Dezhi Zheng, Felix Heide.
International Journal of Computer Vision (IJCV), 2023.
[PDF] [Arxiv] [Zhihu 知乎] 
🌚 Tackling the challenging task of instance segmentation in extreme low-light conditions by introducing a new dataset Low-light Instance Segmentation (LIS) and a learning framework again feature noise.
🎤 Invited Talks
- [2025.08.22] “基于频域分析的视觉表征学习与场景理解” (Visual Representation Learning & Scene Understanding in the Frequency Domain), CCF Forum on Machine Embodied Interaction Intelligence. [Slides] [Event]
- [2025.07.28] “Frequency-aware Feature Fusion for Dense Image Prediction”, Invited Highlight Paper Speaker at the International Conference on Intelligent Computing (ICIC). [Slides] [Event]
- [2025.05.10] “从图像生成到CT图像增强应用” (From Image Generation to CT Image Enhancement Application), 1st Academic Conference on Innovative Development of Smart Health and Intelligent Healthcare (IDAC). [Slides] [Event]
🏅 Awards and Honors
- 2024 China National Scholarship for Doctoral Students, Ministry of Education of the People’s Republic of China
- 2024 CVPR 2024 PBDL International Challenge, Low-Light Track: 2nd Place (Segmentation), 3rd Place (Detection)
- 2024 BMVC 2024 Outstanding Reviewer
- 2022 Beijing Institute of Technology Doctoral Special Scholarship
- 2021 Rocket Force “Intelligent Arrow - Fire Eye” AI Challenge, 7th Place Nationally (Excellence Award)
💬 Professional Service
Journal Reviewer
- IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI)
- IEEE Transactions on Image Processing (TIP)
- International Journal of Computer Vision (IJCV)
- ISPRS Journal of Photogrammetry and Remote Sensing (ISPRS)
- Pattern Recognition (PR)
- Neurocomputing
- Nature Scientific Reports
Conference Reviewer
- IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) 2023, 2024, 2025
- IEEE/CVF International Conference on Computer Vision (ICCV) 2025
- European Conference on Computer Vision (ECCV) 2024
- AAAI Conference on Artificial Intelligence (AAAI) 2025, 2026
- Conference on Neural Information Processing Systems (NeurIPS) 2025, 2026
- British Machine Vision Conference (BMVC) 2023, 2024, 2025
- ACM Multimedia Conference (MM) 2025
- IEEE/CVF Winter Conference on Applications of Computer Vision (WACV) 2026
🧑🏫 Teaching
- Teaching Assistant, Computer Vision, Beijing Institute of Technology, 2021 - 2024 Fall.
📜 Patents
- I have 8 pending national patents, with 3 granted.