This fork is for a tutored research project that we did over the duration of two semesters. The project was part of our syllabus on the IPCVai (Image Processing and Computer Vision ai) EMJM programme, which is a collaboration between UAM (Universidad Autonóma de Madrid), PPCU (Pázmány Páter Catholic University) and UBx (Université de Bordeaux). The project aims to discover the changes in attribute wise performance on the RAP and RAPzs datasets.
The project is based on the excellent works of Jia, Jian and Huang, Houjing and Chen, Xiaotang and Huang, Kaiqi, titled: Rethinking of Pedestrian Attribute Recognition: A Reliable Evaluation under Zero-Shot Pedestrian Identity Setting, (2021 arXiv:2107.03576) and Spatial and semantic consistency regularizations for pedestrian attribute recognition, 2021. (Proceedings of the IEEE/CVF international conference on computer vision, pages: 962-971)
Please check out their REPO for more.
To the staff and lecturers of UAM and PPCU for helping us in our endavors during the Programme.
Special thanks to Juan C. SanMiguel and Álvaro García-Martín for their support in overseeing our research.