About me

Hi! I am an Associate Professor at the University of Beira Interior, Portugal, and a researcher at NOVA LINCS. My primary research focuses on Computer Vision and Biometrics, with particular emphasis on Face Recognition and Deepfake Detection. I am also interested in Explainable AI and in understanding how deep learning models make their decisions.

Latest News

Jul 2026

The project DNXT.AI (PT2030) on intelligent robotic automation for diamond polishing has started — serving as Principal Investigator at UBI.

Apr 2026

My PhD students Cristiano Patrício and Marcelo dos Santos successfully defended their doctoral theses. Congratulations!

Jan 2026

Elected member of the board of the Portuguese Association for Pattern Recognition (APRP) for the 2026–2027 term.

Dec 2025

Paper accepted at AAAI 2026 on fine-grained video anomaly understanding.

Dec 2025

Paper accepted at WACV 2026 on urban waste management.

Dec 2025

Paper accepted at FG 2026 on face morphing attack detection.

Nov 2024

Received the Best Reviewer Award at the British Machine Vision Conference (BMVC 2024).

Oct 2024

Main organizer of RECPAD 2024, the Portuguese national conference on pattern recognition, gathering around 120 researchers.

Selected Publications

AAAI 2026 · CORE A*

FineVAU: A Novel Human-Aligned Benchmark for Fine-Grained Video Anomaly Understanding

J. Pereira, V. Lopes, J. C. Neves, D. Semedo

AAAI Conference on Artificial Intelligence

ACM Computing Surveys 2023 · 99th percentile

Explainable Deep Learning Methods in Medical Diagnosis: A Survey

C. Patrício, L. Teixeira, J. C. Neves

ACM Computing Surveys

IEEE JSTSP 2020 · 94th percentile

GANprintR: Improved Fakes and Evaluation of the State of the Art in Face Manipulation Detection

J. C. Neves, R. Tolosana, R. Vera-Rodriguez, V. Lopes, H. Proença

IEEE Journal of Selected Topics in Signal Processing

IEEE TIFS 2018 · 99th percentile

“A Leopard Cannot Change Its Spots”: Improving Face Recognition Using 3D-Based Caricatures

J. C. Neves, H. Proença

IEEE Transactions on Information Forensics and Security

CVPR 2017 · CORE A*

IRINA: Iris Recognition (even) in Inaccurately Segmented Data

H. Proença, J. C. Neves

IEEE/CVF Conference on Computer Vision and Pattern Recognition

IEEE TPAMI 2016 · 99th percentile

Joint Head Pose/Soft Label Estimation for Human Recognition In-The-Wild

H. Proença, J. C. Neves, S. Barra, T. Marques, J. C. Moreno

IEEE Transactions on Pattern Analysis and Machine Intelligence

All publications →