Gabriel Capiteli Bertocco

I am post-doc researcher in Large-Language Models (LLM) and Computer Vision at the at the Artificial Intelligence Lab., Recod.ai, Institute of Computing, University of Campinas (Unicamp). From 2015 to 2018, I was an undergraduate researcher in a project sponsored by Motorola. The focus was on Deep Learning-based face biometrics analysis for age estimation on mobile devices. For this work, in 2017, I won the Inova Innovation Award for the best undergraduate research project being developed at Unicamp. In 2018, I won First Place in the Face Recognition Competition promoted by the IAPR/IEEE International Summer School in Biometrics and Forensics in Italy. In the same year, I developed a method to detect image repurposing by leveraging Deep Learning techniques, which was the goal of my Final Undergraduate Project. I also worked as a researcher at SciPet Solutions on Technological Innovations in the first semester of 2019. I graduated with distinction in 2019 in Computer Engineering, and finished my Ph.D. in 2024 under supervision of Prof. Dr. Anderson Rocha and Dra. Fernanda Andaló. The Ph.D. research aimed to find coherent groups of identities (people) and things (objects) that appear in a set of potentially non-overlapping camera views in a fully-unsupervised manner. I researched and designed deep learning-based algorithms focused on Unsupervised Person and Objects Re-Identification. The ultimate goal was to design Self-Supervised Learning solutions to handle large unlabeled datasets tackling biases and varied conditions in different image semantics (People, Objects, and Places) and modalities (images and texts). I published the solutions proposed in my Ph.D. in two first-authored publications in the prestigious IEEE Transactions on Information Forensics and Security (T-IFS), which has Impact Factor of 6.8 and the highest h5-index in the Computer Security and Cryptography area as listed by Google Scholar. I have also presented my research and solutions in top-tier conferences such as IEEE Workshop of Information Forensics and Security (WIFS), IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), IEEE Joint Conference on Biometrics (IJCB). The Ph.D. evalution board was composed by Prof. Dr. Patrick Flynn (University of Notre Dame/USA), Prof. Dr. Sébastien Marcel (IDIAP/Switzerland), Prof. Dr. Vitomir ?truc (University of Ljubljana/Slovenia), and Profa. Dra. Esther Luna Colombini (IC/UNICAMP). From 2022 to 2023 I was Professional Research Assistant at University of Colorado Colorado Springs (UCCS/USA) and member of the Biometric Recognition and Identification at Altitude and Range (BRIAR) program (https://www.iarpa.gov/research-programs/briar), a United States Government-supported project devoted to counterterrorism, protection of critical infrastructure, transportation facilities, military force protection, and border security. During this position, I published the designed solutions in IEEE Access, IEEE IJCB 2023 and I've got the third place in the AGReID competition promoted during IEEE IJCB2023. I am also an AI consultant to national and international companies and government institutions, such as, the Police of Dubai, UAE, and UCCS, USA. I have more than eight years of expertise in machine learning, Deep Learning, and Computer Vision with top-tier publications, presentations, consultancies, projects, awards, and one patent. I am also a reviewer for top-tier journals such as IEEE T-IFS and IEEE Transactions on Biometrics, Behavior, and Identity Science. Currently, I am post-doc researcher, focusing on research, innovation and work and consultancy opportunities in AI and Computer Vision. (Source: Lattes Curriculum)