Theodoros Kouzelis
PhD student (NTUA) & Academic fellow
National Technical University of Athens
SHORT BIO
Theodoros holds an MEng in Electrical and Computer Engineering from the National Technical University of Athens. After graduation, he spent two years at the Institute of Language and Speech Processing at Athena RC, working on multiple research projects on automatic speech recognition and generative models for audio and music. Currently, Thodoris is a Ph.D. student at the National Technical University of Athens, working under the supervision of Nikos Komodakis on generative image modeling.
RESEARCH INTERESTS
Thodoris' research interests lie in generative modeling and diffusion models, with a focus on the intersection of representation learning and generative image modeling.
PhD research on "Boosting Generative Image Modeling via Joint Image-feature Synthesis"
Abstract: Latent diffusion models (LDMs) dominate high-quality image generation, yet integrating representation learning with generative modeling remains a challenge. We introduce a novel generative image modeling framework that seamlessly bridges this gap by leveraging a diffusion model to jointly model lowlevel image latents (from a variational autoencoder) and high-level semantic features (from a pretrained self-supervised encoder like DINO).
PhD research on "Boosting Generative Image Modeling via Joint Image-feature Synthesis"
Abstract: Latent diffusion models (LDMs) dominate high-quality image generation, yet integrating representation learning with generative modeling remains a challenge. We introduce a novel generative image modeling framework that seamlessly bridges this gap by leveraging a diffusion model to jointly model lowlevel image latents (from a variational autoencoder) and high-level semantic features (from a pretrained self-supervised encoder like DINO).
