The work carried out in this WP aims at the development of autonomous UAVs to perform visual surveillance in urban environments, based on soft biometric information. The idea is that – upon a security incident – to receive one query composed of a set of soft biometric traits (e.g., “a blonde girl with 4 years, wearing a red shirt and a blue skirt”), and to scan exhaustively one predefined area, to find targets potentially matching the query.
Team & Collaborators
- Hugo Proença | WP leader
- Arun Kumar | Grantee, C4-UBI (06/2019-06/2020)
- Ehsan Yaghoubi | Grantee, C4-UBI
- B. Degardin and H. Proença (2021). Iterative Weak/Self-Supervised Classification Framework for Abnormal Events Detection. In Pattern Recognition Letters, 2021 (to appear).
- E. Yaghoubi, D. Borza, S. V. A. Kumar, and H. Proença (2021). Person Re-identification: Implicitly Defining the Receptive Fields of Deep Learning Classification Frameworks. In Pattern Recognition Letters, 2021 (to appear).
- E. Yaghoubi, S. V. A. Kumar, and H. Proença (2021). SSS-PR: A short survey of surveys in person re-identification. In Pattern Recognition Letters, vol. 143, pp. 50-57, 2021.
- S. V. A. Kumar, E. Yaghoubi, A. Das, B. S. Harish, and H. Proença (2021). The P-DESTRE: A Fully Annotated Dataset for Pedestrian Detection, Tracking, and Short/Long-Term Re-Identification From Aerial Devices. In IEEE Transactions on Information Forensics and Security, vol. 16, pp. 1696-1708, 2021.
- H. Proença, E. Yaghoubi and P. Alirezazadeh (2021). A Quadruplet Loss for Enforcing Semantically Coherent Embeddings in Multi-Output Classification Problems. In IEEE Transactions on Information Forensics and Security, vol. 16, pp. 800-811, 2021.
- International competition on detection and identification of human beings in aerial pictures.
- SCIMAGO special issue is being planned for selected contributions involved in the above competition.