International Conferences
FDSML: Formula-Driven Supervised Metric Learning with Parameter-Aware Triplet Loss
Hina Otake†, Shinei Arakawa†, Keitaro Tanaka†, Yoshihiro Fukuhara, Hirokatsu Kataoka, Shigeo Morishima (†Equal Contribution)
Winter Conference on Applications of Computer Vision (WACV), 2027.
[project page] [code]
Memory Efficient Diffusion Probabilistic Models via Patch-based Generation
Shinei Arakawa, Hideki Tsunashima, Daichi Horita, Keitaro Tanaka, Shigeo Morishima
Generative Models for Computer Vision Workshop in CVPR, 2023.
[paper] [poster] [code]
Domestic Conferences
Formula-Driven Supervised Metric Learning*
Hina Otake†, Shinei Arakawa†, Keitaro Tanaka†, Yoshihiro Fukuhara, Hirokatsu Kataoka, Shigeo Morishima (†Equal Contribution)
Meeting on Image Recognition and Understanding (MIRU) Oral Presentation, August 2026.
Effectiveness of Synthetic Data Pretraining for Generative Models and the Impact of Data Characteristics*
Shinei Arakawa, Keitaro Tanaka, Yoshihiro Fukuhara, Hirokatsu Kataoka, Shigeo Morishima
Meeting on Image Recognition and Understanding (MIRU) Posters, August 2026.
Building a Natural Language Interface for Finite Element Analysis Software Using Generative AI*
Akio Miyoshi, Shinei Arakawa, Shinya Nakamura
The 30th Annual Conference of the Japan Society for Computational Engineering and Science, Keynote Presentation, July 2025.
Any-scale Image Generation via Patch-based Diffusion Models*
Shinei Arakawa, Erik Härkönen, Hideki Tsunashima, Daichi Horita, Shigeo Morishima
Visual Computing (VC) Posters, No. 20, September 2023.
[poster]
Examination of Memory Consumption Reduction in Diffusion Probabilistic Models through Patch Partition*
Shinei Arakawa, Hideki Tsunashima, Daichi Horita, Keitaro Tanaka, Shigeo Morishima
Meeting on Image Recognition and Understanding (MIRU) Posters, July 2023.
[poster]
Diffusion Probabilistic Models for Multi-resolution Image Generation*
Shinei Arakawa, Hideki Tsunashima†, Daichi Horita†, Shigeo Morishima (†Equal Contribution)
85th National Convention in Information Processing Society of Japan (IPSJ), March 2023.
Patch-based Memory Efficient Diffusion Probabilistic Models
Shinei Arakawa, Hideki Tsunashima, Daichi Horita, Keitaro Tanaka, Shigeo Morishima
Visual Computing (VC) Posters, No. 10, October 2022.
* Translated into English from the original Japanese title.
