At IEEE VCIP 2024, Professor Aizawa presented a keynote lecture titled "Building a Realistic Virtual World from 360˚ Videos for Large Scale Urban Exploration."
News
- MOVIE MAP
- FOOD LOG
Track and analyze your daily food choices with our Food Log project!
- MANGA PROJECT
- LECTURES
Links for lectures.
- IEEE Student Branch
VCIP 2024 Keynote Lecture
2024/12
Comic Computing
2024/11
A special session titled "The Journey of Comic Computing from Manga109" was held at Special Interest Group on Comic Computing. The lecture was presented by Aizawa, Baek, Li, and Ikuta from the laboratory.
https://sig-cc.org/?p=623
https://sig-cc.org/?p=623
Accepted for CVIU
2024/11
The following paper was accepted for Computer Vision and Image Understanding.
Q.Yu, G.Irie, K.Aizawa, Open-Set Domain Adaptation with Visual-Language Foundation Models
Q.Yu, G.Irie, K.Aizawa, Open-Set Domain Adaptation with Visual-Language Foundation Models
Outstanding Research Presentation Award at the Institute of Image Information and Television Engineers
2024/11
Takashi Otonari has received the above-mentioned award.
T.Otonari, S.Ikehata, K.Aizawa, 都市シーンにおける動く物体を除去した静的なNeRF表現の学習
T.Otonari, S.Ikehata, K.Aizawa, 都市シーンにおける動く物体を除去した静的なNeRF表現の学習
Accepted for MMM2025
2024/10
The following paper was accepted for Multimedia Modeling 2025.
FoodMLLM-JP: Leveraging Multimodal Large Language Models for Japanese RecipeGeneration, Y. Imajuku, Y. Yamakata, K. Aizawa
FoodMLLM-JP: Leveraging Multimodal Large Language Models for Japanese RecipeGeneration, Y. Imajuku, Y. Yamakata, K. Aizawa
Accepted for NuerIPS WS 2024
2024/10
The following presentations were accepted for NueuIPS WS 2024
- JMMMU: A Japanese Massive Multi-discipline Multimodal Understanding Benchmark, S. Onohara, A. Miyai, Y. Imajuku, K. Egashira, J. Baek, X. Yue, G. Neubig, K. Aizawa
- FlexFlood: Efficiently Updatable Learned Multi-dimensional Index, F. Hidaka, Y. Matsui
Release of Japanese MMMU (JMMMU)
2024/10
We have developed a Japanese extension of a multimodal LLM benchmark. After carefully examining the existing benchmark, MMMU, we translated 24 culturally-neutral subjects into Japanese and created 4 new culturally-dependent subjects, resulting in a total of 1,320 questions (1,118 images). On our website, we also conducted benchmarking of major multimodal LLMs.
https://mmmu-japanese-benchmark.github.io/JMMMU/
https://mmmu-japanese-benchmark.github.io/JMMMU/