Research

My work focuses on the intersection of deep learning and computer vision.

Iron Man body and Spider-Man head reference images condition a generated 3D figurine

SpaceFlow: Locally Controllable 3D Generation

N De La Fuente, J Lafuente Baeza, M Sayfiddinov, F Scharitzer, M Pollefeys, A Çelen, S Deb Sarkar, E Fedele

Preprint · 2026

Training-free 3D generation with local geometric control and part-specific text or image appearance conditioning.

PAH

Prototype Augmented Hypernetworks

CVPR 2025 (LatinX in CV)

Investigating hypernetwork architectures for few-shot learning and model adaptation.

GUIDEX

GUIDEX: Guided Data Generation

ACL 2025 (Findings)

Synthetic data generation strategies for information extraction tasks.

RL Study

Deep RL in Visual Environments

KDD 2024

Evaluating DQN stability across diverse Atari benchmarks.