News
Our paper "Virtual Consistency Model for All-in-one Image Restoration" was accepted by TIP.
Our paper "FreeDehaze: Towards Training-free Real-world Image Dehazing via Diffusion Degradation Prior" was accepted by TIP.
Research Group Attends the 2026 CSIG Young Scientists Conference.
Our Paper "Unifying Heterogeneous Degradations: Uncertainty-Aware Diffusion Bridge Model for All-in-One Image Restoration" Presented at ICML 2026.
Lab Leader Prof. Jin appointed as Young Editorial Board Member of the journal Machine Intelligence Research (MIR).
On November 26, 2025, Professor Zhi Jin delivers a lecture at School of Artificial Intelligence and Robotics, Hunan University.
Our paper "Gradient as Conditions: Rethinking HOG for All-in-one Image Restoration" was accepted by AAAI 2026.
FVL members participated in CSIG 2025, engaging in extensive discussions with numerous experts and scholars.
Our paper "Enhancing Compressed Low-Light Image via Decoupling Strategy and Robust Priors" was accepted by IEEE Transactions on Multimedia.
Our paper "RPD-Diff: Region-Adaptive Physics-Guided Diffusion Model for Visibility Enhancement under Dense and Non-Uniform Haze" was accepted by BMVC 2025.
Our paper "MotionDiff: Training-free Zero-shot Interactive Motion Editing via Flow-assisted Multi-view Diffusion" was accepted by ICCV 2025.
Our paper "Fourier-based Decoupling Networkfor Joint Low-Light lmageEnhancement and Deblurring" was accepted by IEEE Transactions on Image Processing.
FVL members participated in VALSE 2025, engaging in academic exchanges on cutting-edge AI research.
FVL members participated in CCIG 2025 and delivered a paper poster presentation, engaging in extensive discussions with numerous experts and scholars.
Lab Leader Prof. Jin has been promoted to Full Professor in 2025.04.
Our paper "MB-TaylorFormer V2: Improved Multi-branch Linear Transformer Expanded by Taylor Formula for Image Restoration" was accepted by IEEETransactions on Pattern Analysis and Machine Intelligence.
Our paper "Colorization-Inspired Customized Low-Light Image Enhancement by a Decoupled Network" was accepted by IEEE Transactions on Neural Networks and Learning Systems.
Our paper "FDG-Diff Frequency-Domain-Guided Diffusion Framework for Compressed Hazy Image Restoration" was accepted by ICME 2025.
January 22, 2025, members of the Frontier Vision Laboratory participated in the year-end team-building event.
FVL members participated in CSIG 2024 and delivered a paper poster presentation, engaging in extensive discussions with numerous experts and scholars.
December 3, 2024 Professor Jian Zhang, visited our lab and gave a talk.
Our paper "CSPN: A Category-specific Processing Network for Low-light Image Enhancement" was accepted by IEEE Transactions on Circuits and Systems for Video Technology.
November 4, 2024 Professor Chenglin Li, visited our lab and gave a talk.
The 24th master students successfully passed the graduation thesis defense.
FVL members participated in VALSE 2024 and delivered a paper poster presentation, engaging in extensive discussions with numerous experts and scholars.
Students from the Technical University of Munich (TUM) came to Sun Yat-sen University to conduct academic exchanges with the FVL.
Lab leader Prof. Jin with two Ph.D. students went to visit Technical University of Munich for Sino-German Mobility Programme.
We participated in the 9th edition of CVPR2024 workshop NTIRE: New Trends in Image Restoration and Enhancement challenge and won first place among 427 participants.