Convidado
GABRIEL BERTOCCO
CAMPINAS / SPI have been a graduate student pursuing a Ph.D. degree in Computer Science at the Institute of Computing (IC), University of Campinas (Unicamp), Brazil, since August 2019. This research aims to find coherent groups of identities (people), things (objects), and places that appear in a set of potentially non-overlapping camera views in a fully-unsupervised manner. I am studying and deploying Deep Convolutional Neural Networks (DCNNs)-based algorithms focused on Unsupervised Person Re-Identification, extended to objects and places. The ultimate goal is to design Self-Supervised Learning and Clustering techniques to handle large unlabeled datasets tackling biases in different image semantics (People, Objects, and Places) to support investigations and event understanding. This investigation recently led to two first-author publication in the IEEE Transactions on Information Forensics and Security (T-IFS) leveraging Self-Supervised Deep Learning solutions. It was also presented in the IEEE Workshop of Information Forensics and Security (W-IFS) in December 2021, and in the Long-Range Recognition in the Winter Applications on Computer Vision (WACV) Workshop in January of 2023. Those works also inspired our solution for the AGReID2023 Competition in the IEEE International Joint Conference in Biometrics (IJCB2023) achieving the third place in the competition. I have also been part of the Biometric Recognition and Identification at Altitude and Range (BRIAR) program since 2022. It is an US government-sponsored which aims to support missions such as counterterrorism, protection of critical infrastructure and transportation facilities, military force protection, and border security.