Julian Jorge Andrade Guerreiro

I'm currently a PhD student at The University of Tokyo and a member of the Machine Perception Group under the supervision of Professor Hideki Nakayama. My research interests include Computer Vision, Video and Image Processing as well as Image Generation and Editing.

As part of a double master's degree program, I graduated from RWTH Aachen University (supervised by Professor Jens-Rainer Ohm) and Keio University (supervised by Professor Yoshimitsu Aoki). During my internships at Rakuten Institute of Technology and Preferred Networks, I was able to gain valuable experience as a research scientist. I've further had the chance to collaborate on a joint research project with researchers from CyberAgent AI Lab.

Email  /  CV  /  Scholar  /  Twitter  /  Github  /  LinkedIn

profile photo

Research

clean-usnob LayoutFlow: Flow Matching for Layout Generation
Julian Jorge Andrade Guerreiro, Naoto Inoue, Kento Masui, Mayu Otani, Hideki Nakayama
arXiv, 2024
project page / code (coming soon)

Flow Matching for Layout Generation provides a more geometrically intuitive approach, while also significantly speeding up the generation process.

clean-usnob PCT-Net: Full Resolution Image Harmonization Using Pixel-Wise Color Transformations
Julian Jorge Andrade Guerreiro, Mitsuru Nakazawa, Bjorn Stenger
CVPR, 2023
Code

We enable efficient full-resolution Image Harmonization by introducing pixel-wise color transformations (PCTs). Work was done while interning at Rakuten Institute of Technology.

clean-usnob Detection of Osteoarthritis from Multimodal Hand Data
Julian Jorge Andrade Guerreiro, Yoshimitsu Aoki, Shuntaro Saito, Katsuya Suzuki
EMBC, 2022

We investigate the effectiveness of different modalities, namely RGB images, video, and thermal images, to detect Osteoarthritis.


Website design is based on Jon Barron's website .