Browse publications on Google Scholar (top-right) ↗
Resume (English only)
Academic Achievements
Adaptive Vision-Language Prompt Learners for Learning with Noisy Labels (2025)
Dual Polarity Prompts with Stochastic Entropy Perturbation for Label Noise (2025)
VolE++: A Text-Guided Point-cloud Framework for Food 3D Reconstruction and Volume Estimation (2025)
A Cross-Platform, WebGPU-Based 3D Engine for Real-Time Rendering and XR Applications (2025)
Enriching Unbounded Appearances for Neural Radiance Fields (2025)
VolTex: Food Volume Estimation using Text-Guided Segmentation and Neural Surface Reconstruction (2025)
FoodMem: Near Real-time and Precise Food Video Segmentation (2025)
MVSBoost: An Efficient Point Cloud-based 3D Reconstruction (2024)
Decoding Class Dynamics in Learning with Noisy Labels (2024)
Research Experience
Worked as a researcher in real-time global illumination at the University of Minho between 2009 and 2010. Joined Mimetic INRIA research team as a research engineer focusing on Crowd Simulation in 2014. From fall 2016 to summer 2020, worked in the GTI group of Professor Josep Blat at Universitat Pompeu Fabra (Barcelona), first supported by a Marie Sklodowska-Curie individual fellowship on Bayesian Monte Carlo Rendering, then as a Lector. In June 2020, joined the Universitat de Barcelona as a Serra Húnter Tenure Track Lecturer, extending his research activities to deep learning methods in collaboration with Professor Petia Radeva. In fall 2024, joined the Department of Information and Communication Technologies of Pompeu Fabra University to lead its Interactive Technologies Group (GTI) as an Associate Professor.
Education
Received BSc in Informatics from the University of Minho, Portugal in 2007; MSc in Informatics from the same university in 2009. Joined INRIA, France as a PhD student under the supervision of Professor Kadi Bouatouch, in close collaboration with Dr. Christian Bouville and Professor Luís Paulo Santos, from 2010 to 2013.
Background
Research Interests: Photo-Realistic Rendering, Monte Carlo Methods for Global Illumination, Crowd Simulation, Deep Learning with Noisy Labels.