Kai-Fu Yang (杨开富)
Associate Research Professor (副研究员)
Computational Vision, Active Vision, Eye Tracking, Ophthalmic Image Analysis
视觉认知与计算、计算机视觉、眼动跟踪、眼科影像分析

Center for Visual Cognition and Brain-Inspired Computation
University of Electronic Science and Technology of China (UESTC)
No.4, Section 2, North Jianshe Road, Chengdu 610054, China.

Email:yangkf [AT] uestc.edu.cn
[Google Scholar]

About Me

I am an associate research professor at the MOE Key Lab for Neuroinformation, School of Life Science and Technology, University of Electronic Science and Technology of China (UESTC). I received my Ph.D. degree in Biomedical Engineering from UESTC in 2016 under the supervision of Prof. Yong-Jie Li. From August 2019 to August 2020, I was a visiting scholar at the Computer Vision Lab, Department of Information Technology and Electrical Engineering, ETH Zurich, Switzerland.

Research Interests

I conduct interdisciplinary research at the intersection of visual cognition and computer vision. My research aims to explore the underlying computational theory of visual cognition and develop bio-inspired methods for computer vision applications. Through a combination of computational modeling and behavioral experiments such as eye-tracking, we investigate the computational basis of many aspects of vision, including visual perception, active vision, and object recognition.

In addition, we also seek to develop novel artificial intelligence methods for healthcare and human-computer interaction, particularly based on eye-tracking technology and ophthalmic image analysis.

Projects and Publications

(1) | Visual Attention | Object Search | Active Vision |

 

Visual Attention Graph

We propose a new attention representation, called visual attention graph (VAG), to simultaneously code the visual saliency and scanpath in a graph-based representation and better reveal the common attention behavior of human observers. The visual attention graph provides a better benchmark for evaluating attention prediction methods and demonstrates promising potential in assessing human cognitive states.

  • KF Yang* and YJ L*. Visual Attention Graph. Behavior Research Methods, 2026.
    •

     

    Guided Attention and Saliency Detection

  • P Peng, KF Yang*, SQ Liang, YJ Li. Contour-guided Saliency Detection with Long-range Interactions. Neurocomputing, 2022. [Codes]
  • DH He, KF Yang*, XM Wan, F Xiao, HM Yan, YJ Li. A New Representation of Scene Layout Improves Saliency Detection in Traffic Scenes. Expert Syst. Appl.,2022.
  • P Peng, KF Yang, FY Luo, YJ Li. Saliency Detection Inspired by Topological Perception Theory. IJCV, 2021. [Codes]
  • KF Yang, H Li, CY Li, YJ Li. A Unified Framework for Salient Structure Detection by Contour-Guided Visual Search. IEEE TIP, 2016. [Codes]
    •

    (2) | Lightness and Color | Receptive Field | Visual Enhancement |

     

    Gray Anchoring and Gray Pixel [Codes]

    We propose a novel computational theory—Gray Anchoring Theory—to explain how the early visual system contributes to color constancy, and we suggest that concentric double-opponent cells in V1 serve as the neural basis for implementing gray anchoring. Furthermore, we have systematically developed Gray-Pixel methods for illuminant estimation and established a complete technical framework for bio-inspired color constancy by integrating biological mechanisms, computational theories, and algorithmic implementations. We also defined the problem of nighttime color constancy (NCC) and collected the first NCC dataset.

  • KF Yang*, DJ Xing, YJ Li*. Gray Anchoring: a New Computational Theory for Biological Color Constancy. Neuroscience Bulletin, 2026.
  • C Cheng, KF Yang*, XM Wan, LLH Chen, YJ Li. Nighttime Color Constancy Using Robust Gray Pixels. JOSA A, 2024.
  • KF Yang, SB Gao, YJ Li. Efficient Illuminant Estimation for Color Constancy Using Grey Pixels. CVPR, 2015.

    •  

    Lightness and Image Enhancement

  • KF Yang, C Cheng, SX Zhao, HM Yan, XS Zhang, YJ Li. Learning to Adapt to Light. IJCV, 2023. [Codes]
  • KF Yang, XS Zhang, YJ Li. A Biological Vision Inspired Framework for Image Enhancement in Poor Visibility Conditions. IEEE TIP, 2020. [Codes]
  • KF Yang, H Li, HL Kuang, CY Li, YJ Li. An Adaptive Method for Image Dynamic Range Adjustment. IEEE TCSVT, 2019. [Codes]
    •

     

    Contour Detection and Receptive Field Models [Codes]

  • KF Yang, SB Gao, CF Guo, CY Li, YJ Li. Boundary Detection Using Double-Opponency and Spatial Sparseness Constraint. IEEE TIP, 2015.
  • KF Yang, CY Li, YJ Li. Multifeature-based Surround Inhibition Improves Contour Detection in Natural Images. IEEE TIP, 2014.
  • KF Yang, SB Gao, CY Li, YJ Li. Efficient Color Boundary Detection with Color-opponent Mechanisms. CVPR, 2013.


    • (3) | Eye Tracking | Ophthalmic Imaging | AI in Optometry |

     

    Ophthalmic Image Analysis

  • Y Tan, WD Shen,..., KF Yang*, YJ Li*. Retinal Layer Segmentation in OCT images with Boundary Regression and Feature Polarization. IEEE TMI, 2024.
  • Y Tan, SX Zhao, KF Yang*, YJ Li*. A Lightweight Network Guided with Differential Matched Filtering for Retinal Vessel Segmentation. Comput. Biol. Med., 2023.
  • Y Tan, KF Yang*, SX Zhao, YJ Li. Retinal Vessel Segmentation with Skeletal Prior and Contrastive Loss. IEEE TMI, 2022. [Codes]
  • J Wang, YJ Li, KF Yang*. Retinal fundus Image Enhancement with Image Decomposition and Visual Adaptation. Comput. Biol. Med., 2021.
    •

    Professional Activities

    Associate Editor for IET Image Processing.
    Reviewer for IJCV, IEEE T-IP, IEEE T-ITS, IEEE T-CSVT, etc.