Asako Kanezaki
Asako Kanezaki studies 3D recognition and embodied AI. Her work connects visual representations to robot navigation and manipulation. Kanezaki, Jirou Nitta and Yoko Sasaki coauthored GOSELO, a goal-directed obstacle and self-location map for robot navigation using reactive neural networks. The work appeared in 2018. Kanezaki, Wonjik Kim and Masayuki Tanaka coauthored research on unsupervised image segmentation through differentiable feature clustering. It is listed in IEEE Transactions on Image Processing. Kanezaki coauthored the journal version of RotationNet with Yasuyuki Matsushita and Yoshifumi Nishida. The method studies object categorization and unsupervised pose estimation from several views.
Timeline
-
2018
Neural navigation maps
Kanezaki, Jirou Nitta and Yoko Sasaki coauthored GOSELO, a goal-directed obstacle and self-location map for robot navigation using reactive neural networks. Kanezaki, Jirou Nitta and Yoko Sasaki published the work in 2018.
-
2020
Unsupervised image segmentation
Kanezaki, Wonjik Kim and Masayuki Tanaka coauthored research on unsupervised image segmentation through differentiable feature clustering. The research appeared in IEEE Transactions on Image Processing.
-
2021
Joint object recognition and pose
Kanezaki coauthored the journal version of RotationNet with Yasuyuki Matsushita and Yoshifumi Nishida. The method studies object categorization and unsupervised pose estimation from several views.
-
As of October 2026
Embodied AI research team
Kanezaki directs RIKEN’s 3D Environmental Information Understanding Team. Its program uses machine learning for object recognition, mapping, navigation and manipulation.
Sources and credits
Photo: www.riken.jp; Asako Kanezaki profile photograph; creator not stated. License: Not stated by source. Source: riken.jp. Cropped to a square and resized. Ask us to remove or replace this photo.