Creating new ways of seeing the world through perception and intelligence
Technologies for recognizing objects and concepts not seen during training. We pursue AI capable of handling diverse real-world situations, including open-set recognition, zero-shot recognition, and open-vocabulary recognition.
Self-Supervised Learning for Open-Set Scene Graph Generation by Unknown-Object and Relationship Pseudo-Labels
IEEE Access, 2026
Towards Open-Set Scene Graph Generation with Unknown Objects
IEEE Access, 2024
Hierarchical Global-Local Fusion for One-stage Open-vocabulary Temporal Action Detection
ACM TOMM, 2025
We study recognition technologies leveraging information from various sensors, including low-resolution far-infrared sensors and body skeleton sequences. By integrating multiple sensors, we achieve more accurate and robust recognition.
MultiSensor-Home: Multi-modal Multi-view Dataset and Benchmarks for Action Recognition in Home Environments
Pattern Recognition, 2026
Hierarchical Graph Attention Networks with Spatio-Temporal Class Tokens for Distributed Audio-Visual Event Classification
Multimedia Tools and Applications, 2026
LFIR2Pose: Pose Estimation from an Extremely Low-Resolution FIR Image Sequence
ICPR, 2022
Subjective Baggage-Weight Estimation based on Human Walking Behavior
IEEE Access, 2024
We apply computer vision technologies to robotics, tackling challenges such as object-goal navigation. We develop technologies for robots to autonomously perceive their environment and achieve goals.
Category-level Object Pose Estimation in Heavily Cluttered Scenes by Generalized Median Shapes
ICPR, 2024
Best Next-Viewpoint Recommendation by Selecting Minimum Pose Ambiguity for Category-Level Object Pose Estimation
JJSPE, 2021
Median-Shape Representation Learning for Category-Level Object Pose Estimation in Cluttered Environments
IEEE Access, 2023
Projects applying AI technologies to scientific research. In Egyptian archaeology, we use AI for artifact analysis and site surveying. In radio astronomy, we apply AI for advanced data analysis.
3D Survey of the Menkaure Pyramid
ARCE Annual Meeting, 2024
Predicting Reliable H2 Column Density Maps from Molecular Line Data Using Machine Learning
MNRAS, 2023
Development of a High-Speed Identification Model for Infrared-Ring Structures Using CNN
SPIE Astronomical Telescopes, 2022
Scene Understanding and Knowledge Extension by Active Open-world Recognition
JSPS KAKENHI Grant-in-Aid for Scientific Research (A) (FY2024–2028)
AI共同知能による古代エジプト王墓研究の革新
JSPS KAKENHI Grant-in-Aid for Scientific Research (A) (FY2026–2028) / PI: Y. Kawae
星間分子雲の階層構造解析を支えるAI・データ駆動型解析基盤の構築
JHPCN Joint Research Project (FY2026) / PI: Y. Shimajiri
A real-world, human-centered approach to cognition and behavior through collaboration between psychology, AI, and robotics
JSPS KAKENHI Challenging Research (Pioneering) (FY2024–2027) / PI: T. Kumada
International Platform for Building Human Foundation Models towards Omoiyari AI
JST ASPIRE (FY2023–2028) / PI: H. Nishino
人物行動を手掛かりとした車載映像クラウド探索による知識獲得型認識基盤の構築
JSPS KAKENHI Grant-in-Aid for Scientific Research (B) (FY2023–2026) / PI: D. Deguchi
For grants before FY2025, please visit Prof. Kawanishi's personal page .