Paper 'Text-DIAE: A Self-Supervised Degradation Invariant Autoencoders for Text Recognition and Document Enhancement' accepted at AAAI 2023; Participated in organizing the Privacy Preserving Federated Learning Document VQA (PFL-DocVQA) competition at NeurIPS 2023; Received an Excellent mark for his PhD degree from Autonomous University of Barcelona.
Research Experience
Pre-doctoral researcher at CVC, Barcelona, 2019-2022, working on recognizing historical handwritten ciphers; Post-doctoral researcher at CVC, Barcelona, 2022-2023, developing privacy-preserving ML models for document intelligence; Lead AI researcher at Chordata Motion, 2023-2024, developing intelligent motion capture systems; Post-doctoral researcher at CVC, Barcelona, 2024-present, working on explainable, secure, and safe AI systems in the DocVQA field.
Education
Bachelor's degree in Computer Science from University of Monastir, 2012-2015; Master's degree in Computer Science from University of Monastir, 2015-2018; PhD in Computer Science from Autonomous University of Barcelona, 2019-2022, under the supervision of Prof. Andreas Maier (Friedrich-Alexander Universität Erlangen-Nürnberg, Germany), Prof. Ernest Valveny (Universitat Autonoma de Barcelona, Spain), and Dr. Naila Murray (Meta Artificial Intelligence Research, UK).
Background
Research interests include computer vision, machine learning, and document intelligence. Led the machine learning team at Chordata Motion, developing intelligent motion capture/estimation systems.
Miscellany
Email: msouibgui@cvc.uab.cat; Google Scholar Profile; Github Profile; Linkedin Profile