A Learning-based Method for Image Demoiréing

  by   Simeng Qiu






Departments Computer Science
DescriptionThe Moiré effect (https://en.wikipedia.org/wiki/Moir%C3%A9_pattern) is a common phenomenon that frequently exists in our daily life. It occurs as a result of interference between two similar periodic patterns. The Moiré pattern has found various applications in imaging and graphics, such as maximizing the Moiré effect for camera tracking or face-spoofing detection. When photographing in front of a monitor screen, the Moiré effect happens due to the geometric alignment between the LCD panel and the color filter array of the camera. In this project, we aim to minimize the Moiré fringes to improve the quality of recaptured images (from phones or cameras) from the display monitor by applying machine learning techniques.
PreparationThe project would require programming skill using Python or MATLAB. Dataset and source code: https://github.com/CVMI-Lab/UHDM Background: https://en.wikipedia.org/wiki/Moir%C3%A9_pattern
Project Categories Artificial Intelligence (AI), Human Computer Interaction (HCI), Security, Software Engineering, Visual Computing
Project Keywords


Level of Studies

Level 6 (Undergraduate Year 3) yes
Level 7 (Masters) yes
Level 8 (PhD) yes