Masato Ishii
Sony Group Corporation
E-mail: masato0713 [at] gmail [dot] com
English · 日本語
Affiliation: Sony Group Corporation
Research interests: audio and video generation, diffusion models in general
Career
- 2019 – present: Sony Group Corporation
- Seconded to Sony Research Inc. (2023 – 2026)
- 2017 – 2019: RIKEN Center for Advanced Intelligence Project (AIP), Visiting Researcher (concurrent)
- 2010 – 2019: NEC Corporation
- Ph.D. (Information Science and Technology), Graduate School of Information Science and Technology, The University of Tokyo (Advisor: Prof. Masashi Sugiyama)
Recent Publications
Selected recent papers. See Google Scholar for the full list.
- NeurIPSDisen-Forcing: Disentangling Semantic Anchoring from Motion for Autoregressive Video DiffusionIn Advances in Neural Information Processing Systems (NeurIPS), 2026
- ACCVPose Mutation: Target-Pose-Free Consistent Pose Exploration for Portrait GenerationIn Asian Conference on Computer Vision (ACCV), 2026
Invited Talks & Articles
- IBIS 2025 Tutorial: “Diffusion Models: Past and Future” (in Japanese)
- Journal of the Japan Society of Information and Media Studies, July 2025: “Recent Advances in High-Resolution Image/Video Generation with Diffusion Models” (in Japanese)
- IJCNN 2025 Tutorial: “Diffusion Models and Flows” (in English)
- SSII 2024, Research Highlights Session: “The Evolution of Image Generation Technology: The Past Ten Years and Future Prospects” (in Japanese)
- CV Saizensen (Frontiers of Computer Vision), Winter 2023: “An Introduction to Diffusion Models: An Invitation to the New Frontier of Image Generation” (in Japanese)
- FIT 2023 Special Event “Frontiers of Vision-and-Language”: “Diffusion Models in Vision-and-Language” (in Japanese)
- IPSJ Seminar Series 2023: “Image Generation with Diffusion Models and Its Applications” (in Japanese)
- JSAI 2023 (Annual Conference of the Japanese Society for Artificial Intelligence) Tutorial: “Image Generation with Diffusion Models: Fundamentals and Recent Advances” (in Japanese)
- CV Saizensen (Frontiers of Computer Vision), Summer 2023: “Diffusion Models Today: Behind the Magic That Transformed Image Generation” (in Japanese)
- IPSJ SIG-CVIM, May 2023, Tutorial: “Image Generation with Diffusion Models” (in Japanese)
- WEB+DB PRESS Vol. 134, Special Feature 3: “How Image Generation AI Works: Inside Stable Diffusion” (in Japanese)
- IBISML Workshop, March 2023, Invited Talk: “The Frontier of Image Generation with Diffusion Models” (in Japanese)
- SSII 2022, Technology Trend Session: “Machine Learning Techniques for Efficiently Leveraging Limited Data and Labels” (in Japanese)
- GTC Japan 2017, Invited Talk: “Deep Learning Techniques for Extracting the Maximum Value from Data” (in Japanese)
Awards
- Best Paper Award, CVPR 2023 Workshop on AI for Content Creation (AI4CC)
- MIRU Excellent Paper Award, MIRU 2017
- MIRU Nagao Award, MIRU 2016
- Minister of Education, Culture, Sports, Science and Technology Award, the 3rd Technology Management and Innovation Awards
- Best Poster Award, IMPS 2009
- Reviewing: Outstanding Reviewer (ECCV 2026), Gold Reviewer (ICML 2026), Top Reviewer (NeurIPS 2022, 2023), MIRU Reviewer Contribution Award (MIRU 2025)
Professional Service
- Workshop organizer: Gen4AVC (ICCV 2025, ECCV 2026)
- Meeting on Image Recognition and Understanding (MIRU): Executive Chair (2025 – 2026), Finance Chair (2018)
- IEICE Technical Committee on Pattern Recognition and Media Understanding (PRMU): Vice Chair (2025 – 2026), Secretary (2017 – 2018), Committee Member (2016, 2019 – 2022)
- International Conference on Machine Vision Applications (MVA): Publicity Chair (2019), Organizing Committee (2015, 2017)