


Scene recognition is a growing field in computer vision with great potential for behavior understanding and robot localization, among others. In this project, the student is asked to implement deep learning architectures for the recognition of the scene depicted in videos.
1. Glavan A. & Talavera E. (2021). InstaIndoor and Multi-modal Deep Learning for Indoor Scene Recognition. arXiv preprint arXiv:2112.12409.
2. Arnab A. Dehghani M. Heigold G. Sun C. Lučić M. & Schmid C. (2021). Vivit: A video vision transformer. arXiv preprint arXiv:2103.15691.