About
Hello there! I am Hao Chen, and I am from Canton, southeastern China.
Currently, I am a PhD student at the University of Cambridge,
gratefully supervised by Prof. Chao Li and Prof. Carola-Bibiane Schönlieb.
I am also a visiting student at Johns Hopkins University, working with Prof. Zongwei Zhou.
Before join Cambridge, I was at the MI+x Lab
in the University of Birmingham with Prof Jianbo Jiao.
My passion lies in developing better representations by modeling dynamics in images and deepening visual understanding/thinking.
These interests are driven by the goal of tackling challenges in vision and healthcare,
especially in image/video generation and the captivating field of medical imaging.
News
Apr 2026: The MT-PCR won the Best Paper Award in CVM. Congrats Bingxi! 🏆
Mar 2026: Three papers accepted by CVPR 2026 (two highlights), congrats to my collaborators. 🎉
Jan 2026: Two papers accepted by ICLR 2026, congrats Hao Tang for his amazing dedication. 🚀
Jul 2025: One Paper accepted by ICCV 2025, BMVC 2025 and MICCAI 2025 workshop, respectively, congrats to the co-workers. 👏
Apr 2024: Our paper accepted by CVPR 2024 was selected as Oral (3.3%), appreciate the fantastic efforts made by everyone. 🔥
Jul 2023: One paper has been accepted by ICCV 2023. ☕
Oct 2022: Awarded as Kaggle Competition Master (rank top 0.1%). 😎
360+X: A Panoptic Multi-modal Scene Understanding Dataset
Learning Patient-Specific Disease Dynamics With Latent Flow Matching For Longitudinal Imaging Generation
Generalized Spherical Neural Operators: Green’s Function Formulation
Multi-view Self-supervised Disentanglement for General Image Denoising
Scaling Laws in Tumor Segmentation: Best Lessons from Real and Synthetic Data
Exploring Image Representation with Decoupled Classical Visual Descriptors
OG-SAM: Enhancing Multi-Organ Segmentation with Organogenesis-Based Adaptive Modeling
FusDreamer: Label-Efficient Remote Sensing World Model for Multimodal Data Classification
Domain Game: Disentangle Anatomical Feature for Single Domain Generalized Segmentation
Assessing the Performance of Models from the 2022 RSNA Cervical Spine Fracture Detection Competition at a Level I Trauma Center
Transformer-based Multi-label Deep Learning Model for Meticulous Diagnosis of Chronic Ankle Instability: Model Development and External Validation
Expanding from Unilateral to Bilateral: A Robust Deep Learning-based Approach for Radiographic Osteoarthritis Progression
BEVLoc: End-to-end 6-DoF Localization via Cross-modality Correlation Under Bird’S Eye View
Low-light Enhancement in the Frequency Domain
Hierarchical Articulated NeRF for 3D Human Reconstruction and Rendering
Noisy Visual Training: Learning Robust Low-level Image Prior
