Face Understanding

Project Aim: To understand face attributes.

In this project, we work on different aspects of Face Attribute classification, ranging from building state-of-art deep learning systems to understand binary or relative attributes and to semi-supervised learning in the same domain.

Project Team:

Graduate students: Sara Atito Ali Ahmed & Mehmet Can Yavuz

Project Support: TÜBİTAK 1001 (119E429)


  • Mehmet Can Yavuz, Berrin Yanikoglu, 

             YFCC-CelebA Face Attributes Datasets, SIU 2021 (to appear) [pdf

 State-of-art results on relative attribute comparisons.

  State-of-art results on face attribute understanding, on CelebA (%93.20) and LFWA (86.63) datasets

  MTL helps increase performance, in addition to decreasing computational complexity. Pointing to a particular location while learning about a specific attribute helps improve performance and decrease computational complexity.