Wei Shen
Professor
Works on computer vision, including medical image segmentation and 3D shape understanding.
Wei Shen's lab at SJTU works on 3D vision & neural rendering and recognition, detection & segmentation. 73 papers since 2023, 1 at top venues, mostly IJCV. 8 PhD students and 4 master's students. 7 alumni, most often to Huawei and SenseTime. Applications close .
At a glance
- papers, 2023–2026
- 19/ year
- at T1 venues
- 1%
- PhD students
- 8
- MS students
- 4
- alumni listed
- 7
Papers by yearfrom the lab homepage, 2023–2026
- 24 papers2026 · T1 0 · collecting
- 18 papers2025 · T1 1
- 17 papers2024 · T1 0
- 14 papers2023 · T1 0
T1 venuesOtherVenues: IJCV
Research focuspaper titles that use each topic's words
- Recognition, detection & segmentation25 papers
- segmentation20 papers
- 3D vision & neural rendering11 papers
- image segmentation4 papers
- shape0 papers
Paper list
From the lab homepage.
202624 papers
- European Conference on Computer Vision (ECCV), Malmö, SwedenAdaThinking-E: One-Token Entropy Regulation for Adaptive Thinking
- Image Processing, 35: 9661 - 9674BPE-Level Visual-Textual Alignment for Multi-Scene Text Retrieval
- IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Denver, USACodePercept: Code-Grounded Visual STEM Perception for MLLM
- European Conference on Computer Vision (ECCV), Malmö, SwedenCombating Textual Noise and Redundancy: Entropy-Aware Dense Visual Token Pruning
- AAAI Conference on Artificial Intelligence (AAAI), SingaporeCP-CLIP: Customized Parameter Generation for Open-vocabulary Semantic Segmentation
- AAAI Conference on Artificial Intelligence (AAAI), SingaporeDereflection Any Image with Diffusion Priors and Diversified Data
- IEEE Journal of Biomedical and Health Informatic, 30(6): 5167 - 5180Difficulty-adapted Masked Transformer for Semi-Supervised Medical Image Segmentation
- AAAI Conference on Artificial Intelligence (AAAI), SingaporeEfficient Segmentation with Multimodal Large Language Model via Token Routing
- European Conference on Computer Vision (ECCV), Malmö, SwedenExploring Efficient Reasoning Segmentation with Small Language Models
- AAAI Conference on Artificial Intelligence (AAAI), SingaporeFew-step Diffusion for 3D Generation by Learning Marginal Probability Path
- —HyperCLIP++: Fine-tuning CLIP for Open-vocabulary Semantic Segmentation in Hyperbolic Space
- European Conference on Computer Vision (ECCV), Malmö, SwedenLightSTAR: Efficient Visual Document Retrieval via Lightweight Selection with Vision-Adaptive Refinement
202518 papers
- IEEE International Conference on Computer Vision (ICCV), Honolulu, USAA Token-level Text Image Foundation Model for Document Understanding
- Pattern Analysis and Machine Intelligence, 47(5): 3546 - 3562CCDPlus: Towards Accurate Character to Character Distillation for Text Recognition
- IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Nashville, USADomain Generalization in CLIP via Learning with Diverse Text Prompts
- International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI), Daejeon, KoreaEndoDAV: Depth Any Video in Endoscopy with Spatiotemporal Accuracy
- IEEE International Conference on Bioinformatics and Biomedicine (BIBM), Wuhan, ChinaEndoSD-SLAM: Real-time Deformable Endoscopic SLAM via Sparse-Dense Hybrid Representation
- AAAI Conference on Artificial Intelligence (AAAI), Philadelphia, USAFATE: Feature-Adapted Parameter Tuning for Vision-Language Models
- IEEE International Conference on Computer Vision (ICCV), Honolulu, USAGeneralized Tensor-based Parameter-Efficient Fine-Tuning via Lie Group Transformations
- Advances in Neural Information Processing Systems (NeurIPS), San Diego, USAHyperET: Efficient Training in Hyperbolic Space for Multi-modal Large Language Models
- International Conference on Learning Representations (ICLR), SingaporeMaintaining Structural Integrity in Parameter Spaces for Parameter Efficient Fine-tuning
- IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Nashville, USAMarten: Visual Question Answering with Mask Generation for Multi-modal Document Understanding
- Advances in Neural Information Processing Systems (NeurIPS), San Diego, USAOPMapper: Enhancing Open-Vocabulary Semantic Segmentation with Multi-Guidance Information
- IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Nashville, USAParameter-efficient Fine-tuning in Hyperspherical Space for Open-vocabulary Semantic Segmentation
202417 papers
- Graphics (Siggraph Asia)ACM Trans. Graphics (Siggraph Asia)
- European Conference on Computer Vision (ECCV), Milano, ItalyBridging Synthetic and Real Worlds for Pre-training Scene Text Detectors
- Circuits and Systems for Video Technology, 34(2): 1086 - 1097Consensus Synergizes with Memory: A Simple Approach for Anomaly Segmentation in Urban Scenes
- IEEE International Conference on Bioinformatics and Biomedicine (BIBM), Lisbon, PortugalDeCo-Net: Robust Multimodal Brain Tumor Segmentation via Decoupled Complementary Knowledge Distillation Model
- IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Seoul, KoreaDomain-Adaptive Semantic Segmentation Emerges From Vision-Language Supervised Domain-Debiased Self-Training
- Medical Imaging, 43(9): 3211 - 3223Efficient Deformable Tissue Reconstruction via Orthogonal Neural Plane
- International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI), Marrakesh, MoroccoEndoGSLAM: Real-Time Dense Reconstruction and Tracking in Endoscopic Surgeries using Gaussian Splatting
- International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI), Marrakesh, MoroccoMissing as Masking: Arbitrary Cross-modal Feature Reconstruction for Incomplete Multimodal Brain Tumor Segmentation
- IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Seattle, USAParameter Efficient Fine-tuning via Cross Block Orchestration for Segment Anything Model
- AAAI Conference on Artificial Intelligence (AAAI), Vancouver, CanadaPartial Label Learning with a Partner
- European Conference on Computer Vision (ECCV), Milano, ItalyPosFormer: Recognizing Complex Handwritten Mathematical Expression with Position Forest Transformer
- AAAI Conference on Artificial Intelligence (AAAI), Vancouver, CanadaSAM-PARSER: Fine-tuning SAM Efficiently by Parameter Space Reconstruction
202314 papers
- Pattern Analysis and Machine Intelligence, 45(8): 9284 - 9305A Survey on Label-efficient Deep Image Segmentation: Bridging the Gap between Weak Supervision and Dense Prediction
- IEEE International Conference on Computer Vision (ICCV), Paris, FranceAgglomerative Transformer for Human-Object Interaction Detection
- IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Vancouver, CanadaBidirectional Copy-Paste for Semi-Supervised Medical Image Segmentation
- Multimedia, 25: 3737 - 3752ChildPredictor: A Child Face Prediction Framework with Disentangled Learning
- Multimedia, 25: 4267 - 4281Develop then Rival: A Human Vision-Inspired Framework for Superimposed Image Decomposition
- AAAI Conference on Artificial Intelligence (AAAI), Washington, USAIntriguing Findings of Frequency Selection for Image Deblurring
- IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Vancouver, CanadaMagicNet: Semi-Supervised Multi-Organ Segmentation via Magic-Cube Partition and Recovery
- IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Vancouver, CanadaNeRFVS: Neural Radiance Fields for Free View Synthesis via Geometry Scaffolds
- International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI), Vancouver, CanadaNeural LerPlane Representations for Fast 4D Reconstruction of Deformable Tissues
- PDFPattern Analysis and Machine Intelligence, 45(7): 9225 - 9232, 2023
- Advances in Neural Information Processing Systems (NeurIPS), New Orleans, USASegment Anything in 3D with NeRFs
- IEEE International Conference on Computer Vision (ICCV), Paris, FranceSelf-supervised Character-to-Character Distillation for Text Recognition
Where alumni went7 listed, 40% to industry
Industry 2Academia 3
Most often to
- Huawei
- SenseTime
Also to
- JD EXPLORE ACADEMY
- Shanghai University
Apply to SJTUFall 2027
Computer Science PhD (international applicants)
- Round
- Self-funded
- Opens
- —
- GRE
- See program page
- English
- See program page
- Fee
- See program page
Not yet confirmed for this intake.
Similar labsnearest by research summary
- GBUCL
Gabriel Brostow
- 3D vision & neural rendering
- USCMU
Srinivasa G. Narasimhan
- Computational imaging & photography
- SGNTU
Chen Change Loy
- Video understanding
- USUCSD
Nicholas A. Antipa
- Computational imaging & photography
- CNTsinghua
Jie Zhou
- Recognition, detection & segmentation
- USGeorgia Tech
Irfan Essa
- Video understanding
Your notes
Save this lab to keep notes.
Data
Some numbers could not be verified:
- The lab site and the paper record do not agree.
- Collected
- Verified
- Papers
- The lab homepage
- Members
- Lab homepage
- Alumni
- Lab homepage
Tiers are venue classifications, not lab rankings.
SJTUcloses