Rafał Mantiuk

Professor of Graphics and Displays

Department of Computer Science and Technology

The Computer Laboratory

Rainbow Research Group

University of Cambridge

Reserach interest

Applied visual perception; high dynamic range imaging; display algorithms; machine learning for image synthesis; tone-mapping; video coding for new display technologies; image and video quality metrics; visibility metrics; virtual reality and low-level perception; computational photography; computational displays; novel display technologies; colour; perception in computer graphics; novel image and video representations (beyond 2D); psychophysics; modeling visual perception with machine learning.

Biography

Professor of Graphics and Displays, University of Cambridge, Computer Laboratory, UK (from 2023)

Professor/Reader of Graphics and Displays, University of Cambridge, Computer Laboratory, UK (2018-2023)

Senior Lecturer, University of Cambridge, Computer Laboratory, UK (2015-2018)

Lecturer/Senior Lecturer, Bangor University, School of Computer Science, UK (2009-2015)

Postdoc Fellow, University of British Columbia, Canada (2008-2009)

Postdoc, Max-Planck-Institut for Computer Science, Germany (2007-2008)

Internship, Sharp Laboratories of America, Camas WA, USA (2006)

PhD (summa cum laude, Computer Science), Max-Planck-Institut for Computer Science, Germany (2006)

Msc (Computer Science), Technical University of Szczecin, Poland (2003)

Recent projects

Supra-threshold Contrast Perception in Augmented Reality

In optical see-through AR displays, image contrast is much lower than traditional displays due to mixed background light, yet images appear sharper than expected. We explain this effect with a model that describes supra-threshold contrast perception across luminance levels, informing better AR display algorithms and hardware design.

CameraVDP

CameraVDP combines calibrated camera capture with a Visual Difference Predictor to evaluate the visibility of display distortions, such as non-uniformity, color fringing, defective pixels, and others. Our camera calibration employs HDR merging, MTF inversion, vignetting correction, geometric undistortion, homography, and color correction to turn a camera into a measurement instrument.

AR-DAVID

AR-DAVID is a video quality dataset that captures how distortions due to display technologies (e.g, waveguide non-uniformity) are going to be seen on an optical see-through AR display. We found a simple blending of the environment and display light cannot predict the visibility of distortions.

See more projects.

Recent papers

• Resolution limit of the eye - how many pixels can we see?
  Maliha Ashraf, Alexandre Chapiro, and Rafał K. Mantiuk.
  In: Nature Communications, (in print), 2025
  (arxiv) (PDF) (code)
• Supra-threshold Contrast Perception in Augmented Reality
  Dongyeon Kim, Maliha Ashraf, Alexandre Chapiro, Rafał K. Mantiuk.
  In: Proc. of SIGGRAPH Asia 2025, 2025
  (doi) (project page) (PDF) (code)
• CameraVDP: Perceptual display assessment with uncertainty estimation via camera and visual difference prediction
  Yancheng Cai, Robert Wanat, and Rafał K. Mantiuk.
  In: Proc. of SIGGRAPH Asia 2025, 2025
  (doi) (project page) (PDF) (code)
• ColorVideoVDP-ML: Visual Difference Predictor with a neural regressor for image and video quality prediction
  Dounia Hammou, Fei Yin, Rafał K. Mantiuk.
  In: ICME Generalizable HDR & SDR Video Quality Measurement Grand Challenge, 2025
  The second place in the full reference category of the Grand Challenge.
  (project page) (PDF) (code)
• FaceCraft4D: Animated 3D Facial Avatar Generation from a Single Image
  Fei Yin, Mallikarjun B R, Chun-Han Yao, Rafał K. Mantiuk, Varun Jampani.
  In: Proceedings of International Conference on Computer Vision (ICCV'2025), 2025
  (arxiv) (PDF)

See all papers.

Awards and grants

• ColorVideoVDP-ML was awarded second place in the full reference category of ICME Generalizable HDR & SDR Video Quality Measurement Grand Challenge 2025
• HDR-VDP-2 paper received the Test-of-Time Award from SIGGRAPH (2023) for "a significant and lasting impact on computer graphics and interactive techniques over at least a decade"
• HDR-VDP-3 won the 1st place in the HDR Video Quality Measurement Grand Challenge (2023) in the full-reference category
• Color Imaging Conference 2022 (CIC30) Best Paper Award for the paper Suprathreshold contrast matching between different luminance levels
• London Imaging Meeting 2022 Best Paper Award for the paper A Comparative Study on the Loss Functions for Image Enhancement Networks
• ACM SIGGRAPH European Conference on Visual Media Production (CVMP 2022) Best Paper Award for the paper Distilling Style from Image Pairs for Global Forward and Inverse Tone Mapping
• Color Imaging Conference (CIC28) Best Paper Award for the paper Practical color contrast sensitivity functions for luminance levels up to 10000 cd/m²
• Electronic Imaging / Human Vision and Electronic Imaging 2020 Best Paper Award for the paper Predicting visible flicker in temporally changing images
• The IEEE Virtual Reality 2019 Best Journal Paper - for the paper Temporal Resolution Multiplexing: Exploiting the limitations of spatio-temporal vision for more efficient VR rendering
• Electronic Imaging / Human Vision and Electronic Imaging 2019 Best Paper Award for the paper A visual model for predicting chromatic banding artifacts
• ERC Consolidator Grant (2017) - EyeCode: Perceptual encoding of high fidelity light fields
• MSCA Innovative Training Network (2018) - RealVision: Hyper-realistic Visual Experience
• EPSRC research grant (2017) - A spatio-chromatic colour appearance model for retargeting high-dynamic-range image appearance across viewing conditions
• HPC Wales Research and Innovation grant (2013/14) - Video retargeting for delivery to mobile and future display technologies
• Royal Society Research Grant (2013) - Limiting factors of perceptual image fidelity
• EPSRC grant EP/I006575/1 (2011) - Quantifying image quality in computer graphics
• Heinz Billing Award 2006

My contribution to the organization of research networks and conferences can be found here.