skip to content

Researcher: Angelica I. Aviles-Rivero

In real-world scenes, the conditions are often far from optimal, and photographs/videos have to be taken in which target objects are covered by atefacts appear in undesired places for example reflections or due bad weather conditions such as rain. This does not only affect amateur photography; such artefacts may also arise in documentation in museums and aquariums, or black-box cameras in cars, surveillance systems, among others. It is therefore unsurprising that the problem of how to remove reflection artefacts is of great interest, from both practical and theoretical points of view.

Jointly work with: Yale-NUS, NUS, Microsoft+AI, Shandong University and EPFL.

Selected Related Papers:

R. Li, RT Tan, A.I Aviles-Rivero, Q. Fan, C-B Schönlieb and LF Cheong. RainFlow: Optical Flow under Rain Streaks and Rain Veiling Effect. IEEE/CVF International Conference on Computer Vision (ICCV), 2019.

[Tutorial] Asian Conference on Computer Vision ACCV, Perth Australia, 2018.
Classic vs Deep Vision. What is Beyond Deep Learning in Computer Vision? A.I Aviles-Rivero (University of Cambridge), CB. Schonlieb (University of Cambridge),  R. Tan (Yale-NUS), S. You (Data61/ ANU), Q. Fan (Shandong University) and SM Alsaleh (GWU).  LINK:

Related Publications 

Mirror, Mirror, on the Wall, Who’s Got the Clearest Image of Them All?—A Tailored Approach to Single Image Reflection Removal
D Heydecker, G Maierhofer, AI Aviles-Rivero, Q Fan, D Chen, C-B Schonlieb, S Susstrunk – IEEE Transactions on Image Processing (2019) 28, 6185