Xin Li 李鑫 I am currently a Ph.D. Candidate in MOE-Microsoft Key Laboratory of Multimedia Computing and Communication, University of Science and Technology of China (USTC), supervised by Zhibo Chen. Prior to that, I obtained my B.S. degree in Department of Electronic Engineering and Information Science from the University of Science and Technology of China (USTC). From Apr. 2021 to Apr. 2022, I was a research intern at Intelligence Media Group of MSRA under the supervision of Cuiling Lan and Wenjun Zeng. And now, I am a visiting Ph.D. of Learning and Vision (LV) Lab at the National University of Singapore under the supervision of Xinchao Wang. My research interests mainly focus on image/video processing, computer vision, and machine learning, causal learning. Email: lixin666@mail.ustc.edu.cn  /  Google Scholar  /  GitHub

• Latest Works!!!:
  • FreqAlign: Excavating Perception-oriented Transferability for Blind Image Quality Assessment from A Frequency Perspective. (IEEE TMM 2023)
    Xin Li, Yiting Lu, Zhibo Chen.
  • GraphAdapter: Tuning Vision-Language Models With Dual Knowledge Graph . (NeurIPS2023)
    Xin Li, Dongze Lian, Zhihe Lu, Jiawang Bai, Zhibo Chen*, Xinchao Wang*.
  • Diffusion Models for Image Restoration and Enhancement - A Comprehensive Survey. (Arxiv2023)
    Xin Li, Yulin Ren, Xin Jin, Cuiling Lan, Xingrui Wang, Wenjun Zeng, Xinchao Wang, Zhibo Chen.
  • Learning Distortion Invaraint Representation for Image Restoration from A Causality Perspective. (CVPR2023)
    Xin Li, Bingchen Li, Xin Jin, Cuiling Lan, Zhibo Chen*.
  Code Release
  • Enhance-Anything with the pouplar Segment-anything (SAM) is started to be developed and released.

• Image/Video Compression:
  • Task-driven Semantic Coding via Reinforcement Learning. (IEEE TIP2021)
    Xin Li, Jun Shi, Zhibo Chen*.
  Code Release
  • Learned Block-based Hybrid Image Compression. (IEEE TCSVT2021)
    Yaojun Wu#, Xin Li# (Equal Contribution)#, Zhizheng Zhang, Xin Jin, Zhibo Chen*.
  • Hierarchical Reinforcement Learning Based Video Semantic Coding for Segmentation. (IEEE VCIP2022)
    Guangqi Xie, Xin Li, Shiqi Lin, et al.
  • Learned Lossless JPEG Transcoding via Joint Lossy and Residual Compression (IEEE VCIP2022)
    Xiaoshuai Fan, Xin Li, Zhibo Chen*.

• Image/Video Processing/Restoration:
  • Few-Shot Real Image Restoration via Distortion-Relation Guided Transfer Learning. (Preprint)
    Xin Li, Xin Jin, Jun Fu, Xiaoyuan Yu, Bei Tong, Zhibo Chen*.
  • HST: Hierarchical Swin Transformer for Compressed Image Super-resolution. (ECCV2022-workshop)
    Bingchen Li, Xin Li, Yiting Lu, et. al..
  Code Release
  • Learning Omni-frequency Region-adaptive Representations for Real Image Super-Resolution. (AAAI-2021)
    Xin Li, Xin Jin, Tao Yu, Yingxue Pang, Simeng Sun, Zhizheng Zhang, Zhibo Chen*.
  • Dual Prior Learning for Blind and Blended Image Restoration. (TIP2021)
    Xin Jin, Li Zhang, Chaowei Shan, Xin Li, Zhibo Chen*

  • Learning Disentangled Feature Representation for Hybrid-distorted Image Restoration. (ECCV-2020)
    Xin Li, Xin Jin, Jianxin Lin, Sen Liu, Yaojun Wu, Tao Yu, WeiZhou, Zhibo Chen*.  
  • Multi-scale Grouped Dense Network for VVC Intra Coding. (CVPR-Workshop2020)
    Xin Li, Simeng Sun, Zhizheng Zhang, Zhibo Chen*.
  • FAN: Frequency Aggregation Network for Real Image Super-resolution. (ECCV-Workshop 2020)
    Yingxue Pang#, Xin Li#, Xin Jin, Yaojun Wu, Jianzhao Liu, Sen Liu, Zhibo Chen*.
  （# indicates equal contribution）

• Image/Video Quality Assessment:
  • SwinIQA: Learned Swin Distance for Compressed Image Quality Assessment. (CVPR-Workshop2022)
    Jianzhao Liu, Xin Li, Yanding Peng, Tao Yu, Zhibo Chen*.
  Code Release
  • RTN: Reinforced Transformer Network for Coronary CT Angiography Vessel-level Image Quality Assessment. (MICCAI2022)
    Yiting Lu, Jun Fu, Xin Li, et al.

• Image/Video Quality Assessment & Generalization:
  • StyleAM: Perception-Oriented Unsupervised Domain Adaption for Non-reference Image Quality Assessment. (Arxiv version)
    Yiting Lu#, Xin Li# (Eqaul contribution), Jianzhao Liu, Zhibo Chen*
  • LIQA: Lifelong Blind Image Quality Assessment. (TMM2022)
    Jianzhao Liu, Wei Zhou, Xin Li, Jiahua Xu, Shukun An, Zhibo Chen*

  • Source-free Unsupervised Domain Adaptation for Blind Image Quality Assessment. (Arxiv version)
    Jianzhao Liu#, Xin Li# (Equal Contribution), Shukun An, Zhibo Chen*

• Image/Video Processing & Generalization
  • LIRA: Lifelong Image Restoration from Unknown Blended Distortions. (ECCV2020)
    JianZhao Liu, Jianxin Lin, Xin Li, WeiZhou, Sen Liu, Zhibo Chen*.

• Causal Learning & Generalization & Data augmentation
  • Confounder Identification-free Causal Visual Feature Learning. (Arxiv version)
    Xin Li, Zhizheng Zhang*, Guoqiang We, Cuiling Lan*, Wenjun Zeng, Xin Jin, Zhibo Chen*.
  • SelectAugment: Hierarchical Deterministic Sample Selection for Data Augmentation. (AAAI2023 Oral)
    Shiqi Lin, ZhiZheng Zhang, Xin Li, Wenjun Zeng, Zhibo Chen*.

Academic Services

• Invited as a reviewer/PC member for conferences, including ECCV2024, ICML2024, CVPR2024, ICLR2024, AAAI2024, NeurIPS2023, ICCV2023, ICLR2023 workshop, CVPR2023, AAAI2023, CVPR2022, ECCV2022, VCIP2020, VCIP2021, VCIP2022, and journals, including TIP, TNNLS, TMM, TCSVT, JSAC.

Competitions

• The first prize in the 5th Workshop and Challenge on Learned Image Compression (Perceptual Metric Track) (CVPR-CLIC2022).
• The third prize in the 2019 4K+HDR National Artificial Intelligence Competition.(3/1000+)
• The third prize in the NTIRE2020 video deblurring task.

Awards

• Bytedance Scholarship 2022 (Ten students in China)
• China Scholarship Council (CSC) scholarship
• China National Scholarship for Graduate Student (2021)
• NeurIPS Scholar Award, 2023

Research Experience

• Sept. 2019 - Present: Ph.D. Candidate, MOE-Microsoft Key Laboratory of Multimedia Computing and Communication, Mentor: Zhibo Chen.
• Apr. 2021 - Apr. 2022: Research Intern, Intelligent Multimedia Group, Microsoft Research Asia, Mentor: Cuiling Lan and Wenjun Zeng.
• Jan. 2023 - ： CSC Visiting Ph.D., LV label, NUS, Mentor: Xinchao Wang.