Welcome to the home of the Vision & InterAction Group (Vintage)

Led by Dr. Stefan Winkler, Vintage's research focuses on perception-based computer vision, image processing and video analysis.

Vintage is now part of the School of Computing at the National University of Singapore (NUS). The group has its origins at the Advanced Digital Sciences Center (ADSC), the Singapore-based research center of the University of Illinois at Urbana-Champaign (UIUC), with funding from Singapore's Agency for Science, Technology and Research (A*STAR) and National Research Foundation (NRF).

News & Events

[10-Aug-2021] Chen Su, Kang Wangkai, Liu Chaojie and Foo Chi Fa join us for their final year projects (FYP).

[26-Jul-2021] Giving a keynote talk on "AI with EQ - Analyzing Human Emotions" at the 44th International Conference on Telecommunications and Signal Processing (TSP).

[19-Jul-2021] Daniela Winter from University of Tübingen joins us for an internship.

[24-Jun-2021] Invited to a panel on "Visual information processing and machine learning at the crossroads" at 9th European Workshop on Visual Information Processing (EUVIP).

[10-May-2021] Manish Sridhar, Soumyadeep Roy, Meetal Manke, and Aakash Khandelwal join us for their internships.

[29-Apr-2021] We're organizing a Workshop on Synthetic Multimedia - Audiovisual Deepfake Generation and Detection in conjunction with the 29th ACM International Conference on Multimedia 2021.

[27-Jan-2021] Qilin Li joins us for a semester project.

[11-Jan-2021] Calvin Chen Xingzhu joins us for his Final Year Project.

[30-Nov-2020] Aakanksha Chouhan joins us for an internship.

[18-Nov-2020] Stefan Winkler was invited speaker and panelist for the Special Industry Track at IEEE TENCON 2020.

[2-Nov-2020] Our FaceScrub Dataset has been downloaded more than 10,000 times since we released it 6 years ago!

[15-Oct-2020] Our ICLR 2019 paper on The unusual effectiveness of averaging in GAN training was featured in deeplearning.ai's new GAN specialization on coursera, specifically in this video.

[7-Oct-2020] We have released HelipadCat, a new annotated dataset of helipad images of different categories.