HSHiromitsu Shiina
Hiromitsu Shiina
*This is an algorithmically generated author record based on publicly available information.
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Hiromitsu Shiina has a total of 4 co-authors. Here are the top co-authors by number of shared publications:
- Chonho Lee
Department of Information Science and Engineering, Okayama University of Science, Okayama 700-0005, Okayama, Japan
1 shared publication - Haruno Fusa
Department of Information Science and Engineering, Okayama University of Science, Okayama 700-0005, Okayama, Japan
1 shared publication - Kanshin Fusa
Department of Communication and Information System Program, National Institute of Technology Tsuyama College, Tsuyama 708-8509, Okayama, Japan
1 shared publication - Sakuei Onishi
Department of Information Science and Engineering, Okayama University of Science, Okayama 700-0005, Okayama, Japan
1 shared publication
Latest Publications
- Open Access
- 2 February 2026
- Published by MDPI AG in Information
- Vol. 17 (2) , 149
Template-Based Evaluation of Stable Diffusion via Attention Maps
Text-to-image models such as Stable Diffusion (SD) require comprehensive, fine-grained, and high-precision methods for evaluating text–image alignment. A prior method, the text–image alignment metric (TIAM), employs a template-based approach for fine-grained, high-precision evaluation; however, it is restricted to objects and colors, limiting its comprehensiveness. This study extends the TIAM by incorporating attention maps and vision–language models to deliver a fine-grained and high-precision evaluation framework that goes beyond colors and objects to include attributes, actions, and positions. In our experiments, we analyze the evaluation scores of images generated by the proposed method and compare them with human judgments. The results demonstrate that the proposed method outperforms existing methods, exhibiting a stronger correlation with human judgments (r = 0.853, ). In addition, we applied the proposed method to evaluate the generation abilities of three SD models (i.e., SD1.4, SD2, and SD3.5). Each experiment used over 900 images, totaling 9858 images across all experiments to ensure statistical significance. The results indicate that SD3.5 exhibits superior expressiveness compared with SD1.4 and SD2. Nevertheless, for more complex tasks such as multi-attribute generation or multi-action generation, limitations in text–image alignment remain evident.- 2 February 2026
- Published by MDPI AG in Information
- Vol. 17 (2) , 149
Open AccessTemplate-Based Evaluation of Stable Diffusion via Attention Maps
Text-to-image models such as Stable Diffusion (SD) require comprehensive, fine-grained, and high-precision methods for evaluating text–image alignment. A prior method, the text–image alignment metric (TIAM), employs a template-based approach for fine-grained, high-precision evaluation; however, it is restricted to objects and colors, limiting its comprehensiveness. This study extends the TIAM by incorporating attention maps and vision–language models to deliver a fine-grained and high-precision evaluation framework that goes beyond colors and objects to include attributes, actions, and positions. In our experiments, we analyze the evaluation scores of images generated by the proposed method and compare them with human judgments. The results demonstrate that the proposed method outperforms existing methods, exhibiting a stronger correlation with human judgments (r = 0.853, ). In addition, we applied the proposed method to evaluate the generation abilities of three SD models (i.e., SD1.4, SD2, and SD3.5). Each experiment used over 900 images, totaling 9858 images across all experiments to ensure statistical significance. The results indicate that SD3.5 exhibits superior expressiveness compared with SD1.4 and SD2. Nevertheless, for more complex tasks such as multi-attribute generation or multi-action generation, limitations in text–image alignment remain evident.