Naoto Inoue

I am Naoto Inoue (井上 直人).

I am working as a Senior Machine Learning Engineer at Apple in Tokyo, Japan.

Previously, I was a Research Scientist at CyberAgent Inc. AILab in Tokyo, Japan. I was working on multimodal/multitask generative models for automatic visual advertisement creation (e.g., banners and posters) via generating images and texts and compositing them. I got my Ph.D. in 2021 at The University of Tokyo under supervision of Prof. Toshihiko Yamasaki working on GenAI. I was fortunate to collaborate with Adobe Research during my Ph.D. study.

Email  /  CV  /  Google Scholar  /  Twitter  /  Github  /  Linkedin

profile photo
Recent news (selected)

[Apr. 2025] After four wonderful years at CyberAgent AILab, I have started to work as Senior Machine Learning Engineer at Apple (remotely in Tokyo, Japan).
[Feb 2025] One paper is accepted to CVPR2025.
[July 2024] One paper is accepted to ECCV2024.
[Feb. 2024] One paper is accepted to CVPR2024. We'll also host a Workshop on Graphic Design Understanding and Generation (GDUG) at CVPR2024.
[Feb. 2023] Two first-authored papers are accepted to CVPR2023.
[Apr. 2021] I passed the defense and got a Ph.D. degree at The University of Tokyo! I started to work as a Research Scientist at CyberAgent Inc. AILab (in Tokyo, Japan).

Projects
Workshop on Graphic Design Understanding and Generation (GDUG)
Kota Yamaguchi, Yizhi Wang, Naoto Inoue, Mayu Otani, Xueting Wang,
CVPR Workshops, in conjunction with CVPR2024
project
Preprints
Multimodal Markup Document Models for Graphic Design Completion
Kotaro Kikuchi, Naoto Inoue, Mayu Otani, Edgar Simo-Serra, Kota Yamaguchi
arXiv 2024
project / paper / code
LTSim: Layout Transportation-based Similarity Measure for Evaluating Layout Generation
Mayu Otani, Naoto Inoue, Kotaro Kikuchi, Riku Togashi
arxiv 2024
paper
Publications
Type-R: Automatically Retouching Typos for Text-to-Image Generation
Wataru Shimoda, Naoto Inoue, Daichi Haraguchi, Hayato Mitani, Seiichi Uchida, Kota Yamaguchi
CVPR 2025 highlight
project / paper / code
Can GPT Evaluate Graphic Designs Based on Design Principles?
Daichi Haraguchi, Naoto Inoue, Wataru Shimoda, Hayato Mitani, Seiichi Uchida, Kota Yamaguchi
SIGGRAPH ASIA technical communications 2024
project / paper / code
LayoutFlow: Flow Matching for Layout Generation
Julian Jorge Andrade Guerreiro, Naoto Inoue, Kento Masui, Mayu Otani, Hideki Nakayama
ECCV 2024
project / paper / code
OpenCOLE: Towards Reproducible Automatic Graphic Design Generation
Naoto Inoue*, Kento Masui*, Wataru Shimoda*, Kota Yamaguchi (*: equal contribution)
CVPRW (GDUG) 2024, extended abstract
paper / code
Retrieval-Augmented Layout Transformer for Content-Aware Layout Generation
Daichi Horita, Naoto Inoue, Kotaro Kikuchi, Kota Yamaguchi, Kiyoharu Aizawa
CVPR 2024 oral
project / paper / code
Towards Flexible Multi-modal Document Models
Naoto Inoue, Kotaro Kikuchi, Edgar Simo-Serra, Mayu Otani, Kota Yamaguchi
CVPR 2023 highlight
project / paper / code
LayoutDM: Discrete Diffusion Model for Controllable Layout Generation
Naoto Inoue, Kotaro Kikuchi, Edgar Simo-Serra, Mayu Otani, Kota Yamaguchi
CVPR 2023
project / paper / code
Generative Colorization of Structured Mobile Web Pages
Kotaro Kikuchi, Naoto Inoue, Mayu Otani, Edgar Simo-Serra, Kota Yamaguchi
WACV 2023
paper / code & dataset
Learning from Synthetic Shadows for Shadow Detection and Removal
Naoto Inoue, Toshihiko Yamasaki
IEEE TCSVT 2021
paper / code & dataset
RGB2AO: Ambient Occlusion Generation from RGB Images
Naoto Inoue, Daichi Ito, Yannick Hold-Geoffroy, Long Mai, Brian Price, Toshihiko Yamasaki
CGF (proc. of Eurographics) 2020
paper / supp. / video
Learning to Trace: Expressive Line Drawing Generation from Photographs
Naoto Inoue, Daichi Ito, Ning Xu, Jimei Yang, Brian Price, Toshihiko Yamasaki
CGF (proc. of Pacific Graphics) 2019
paper / supp. / video
Fully Convolutional Network with Multi-Step Reinforcement Learning for Image Processing
Ryosuke Furuta, Naoto Inoue, Toshihiko Yamasaki
AAAI 2019
project / paper
Cross-Domain Weakly-Supervised Object Detection through Progressive Domain Adaptation
Naoto Inoue, Ryosuke Furuta, Toshihiko Yamasaki, Kiyoharu Aizawa
CVPR 2018
project / paper

Design: jonbarron