Introduction

Hi, my name is Luiz Guilherme, but everyone calls me Luiz. I enjoy applying computing projects to computer vision, which has been my research focus for the past few years. I am particularly interested in leveraging advanced machine learning techniques to solve real-world challenges in this field.

I like playing games, although I haven’t had much time recently. My top five favorite games (not in order, as I haven’t decided yet) are The Witcher 3: Wild Hunt, The Elder Scrolls V: Skyrim, Red Dead Redemption 2, Elden Ring, and Baldur’s Gate 3. I enjoy open-world games with intriguing stories, and most of these are titles I’ve played multiple times, appreciating narratives.

Nowadays, I’m more focused on watching animes and TV series. You can check out my Letterboxd profile, where I try to log every movie I watch, though I often forget. So far, I’ve recorded 334 movies that I can remember watching.

Biography

In 2024, I began my PhD in Engineering at École de Technologie Supérieure (ÉTS), under the guidance of Prof. Alessandro Koerich and Prof. Éric Granger. My research focuses on Multimodal Learning for Emotion Recognition, exploring approaches to combining modalities such as audio and text while addressing challenges related to missing data and noisy signals.

During my master’s degree in Electrical Engineering at the Universidade Politécnica de São Paulo (USP), I focused entirely on research, refining my machine learning skills and specializing in machine learning for medical imaging. Under the supervision of Prof. Fátima de Lourdes dos Santos Nunes, I presented my work internationally and published more than five papers as the first author.

I earned my Bachelor’s degree in Computer Engineering from USP, where I had the opportunity to develop my skills through diverse experiences. I participated in competitive programming extension groups, began studying machine learning in depth, and gained professional experience through internships and working at three different companies.

Publications

• Enhancing Dental Caries Classification in CBCT Images by using Image Processing and Self-Supervised Learning
  Luiz Zanini, Fátima Nunes, and Izabel Rubira-Bullen
  Computers in Biology and Medicine, Published online Oct 7, 2024. DOI: https://doi.org/10.1016/j.compbiomed.2024.109221
• A Systematic Review on Caries Detection, Classification, and Segmentation from X-Ray Images: Methods, Datasets, Evaluation, and Open Opportunities
  Luiz Zanini, Fátima Nunes, and Izabel Rubira-Bullen
  Journal Imaging Inform Med. Published online March 1, 2024. DOI: https://doi.org/10.1007/s10278-024-01054-5
• Segmentation and Classification of Dental Caries in Cone Beam Tomography Images using Machine Learning and Image Processing
  Luiz Zanini, Fátima Nunes, and Izabel Rubira-Bullen
  In Proceedings of the 17th International Joint Conference on Biomedical Engineering Systems and Technologies - HEALTHINF, pages 428-435. DOI: https://doi.org/10.5220/0012365300003657
• Identification and quantification of caries from CBCT segmented images
  Luiz Zanini, Fátima Nunes, and Izabel Rubira-Bullen
  In Proceedings of the 23rd Brazilian Symposium on Computing Applied to Health, June 27, 2023, São Paulo/SP, Brazil. SBC, Brasil, 1-12. DOI: https://doi.org/10.5753/sbcas.2023.229376
• Convolutional architectures with LSTM and TCN to embolism classification: exploring dependency between data
  Luiz Zanini, Aldomar Silva, Felipe Almeida, Fátima Nunes, and Anna Costa
  In Proceedings of the 19th National Meeting on Artificial and Computational Intelligence, November 28, 2022, Brazil. SBC, Brasil, 461-472. DOI: https://doi.org/10.5753/eniac.2022.227585

Honors

• Best Final Paper of the Computer Engineering Poli - USP 2022
  The final project involved using image processing for segmentation and machine learning to classify the intensity of caries, as well as incorporating virtual reality for result visualization.
• Second place for best paper in the undergraduate category at the National Meeting of Artificial and Computational Intelligence (ENIAC)
  The research focused on image classification in computed tomography for pulmonary embolism classification, leveraging convolutional architectures with LSTM and TCN modules.