Close

INESC TEC

ML Research Intern

Back to Portfolio

INESC TEC

ML Research Intern

Duration: March 2025 - July 2025

Location: Porto, Portugal

Department: Computer Vision and Machine Learning

Overview

Conducted research under a scientific initiation grant focusing on explainable facial recognition solutions. This internship was part of my Master's thesis research and contributed to advancing the field of interpretable AI in computer vision.

Key Responsibilities

  • Developed and implemented explainable AI algorithms for facial recognition systems
  • Conducted extensive literature review on state-of-the-art explainable AI techniques
  • Collaborated with research team on novel approaches to facial recognition transparency
  • Prepared and submitted research findings to an international conference
  • Documented research methodology and results for academic publication

Key Achievements

  • Published Research: Successfully published a scientific article to ICCV (International Conference on Computer Vision)
  • Technical Innovation: Contributed to advancing transparency in AI facial recognition systems with the introduction of two innovative explainable AI solutions for facial recognition: MaskSiamese and ProtoSiamese
  • Academic Recognition: Research work recognized with a final dissertation grade of 19/20

Technologies & Tools Used

Python PyTorch TensorFlow NumPy

Research Impact

The research conducted during this internship has contributed to the broader understanding of explainable AI in facial recognition systems. The published work at ICCV represents a significant contribution to the field and demonstrates the potential for creating more transparent and trustworthy AI systems in computer vision applications.

Skills Developed

  • Advanced machine learning and deep learning techniques
  • Explainable AI and interpretable machine learning
  • Computer vision and image processing
  • Research methodology and academic writing
  • Scientific paper preparation and publication
  • Collaboration in research environments
  • Critical analysis of state-of-the-art literature
Back to Portfolio