Hidenobu Matsuki

Hide Matsuki

I am currently pursuing my PhD at the Dyson Robotics Lab at Imperial College London, under the supervision of Prof. Andrew Davison.
During my PhD, I was a research intern at Google Zurich hosted by Federico Tombari and Keisuke Tateno.

Previously, I completed my Master's degree at the University of Tokyo. During my master's program, I had the opportunity to work at the Technische Universität München (TUM) with Prof. Daniel Cremers.

Between my master's and PhD degree, I worked as a computer vision engineer at Artisense, a startup co-founded by Prof. Cremers.

Email / Scholar / Twitter / Github

Research

My research focuses on real-time computer vision and robotics, with a particular emphasis on visual SLAM.

I strongly believe all the SLAM innovations are made in the process of pursuing hard realtime operation and live interaction (e.g. tracking/mapping seperation, keyframing, scene representations, Gauss-Newton variants, tools like Sophus/Pangolin). To this end, I work on developing efficient algorithms and 3D representations for SLAM, tailored to real-world, real-time scenarios.

Publications

	Gaussian Splatting SLAM Hidenobu Matsuki, Riku Murai, Paul H.J. Kelly, Andrew J. Davison IEEE / CVF Computer Vision and Pattern Recognition Conference (CVPR), 2024 (Highlight, Best Demo Award) project page / video / paper The first Monocular SLAM using Gaussian Splatting as a master representation. The method also supports RGB-D mode.
	NEWTON: Neural View-Centric Mapping for On-the-Fly Large-Scale SLAM Hidenobu Matsuki, Keisuke Tateno, Michael Niemeyer, Federico Tombari, arXiv, 2023 paper Multiple local Neural Fields for efficiently handling loop closure in large-scale SLAM.
	iMODE:Real-Time Incremental Monocular Dense Mapping Using Neural Field Hidenobu Matsuki, Edgar Sucar, Tristan Laidlow, Kentaro Wada, Raluca Scona, Andrew J. Davison IEEE International Conference on Robotics and Automation (ICRA), 2023 (Oral Presentation, Best Navigation Paper Award Finalist) paper Generic real-time monocular dense mapping pipeline utilizing the smoothness prior inhereint in Multi Layer Perceptron.
	CodeMapping: Real-Time Dense Mapping for Sparse SLAM using Compact Scene Representations Hidenobu Matsuki, Raluca Scona, Jan Czarnowski, Andrew J. Davison, IEEE Robotics and Automation Letters (RA-L), 2021 paper / video Real-time dense mapping framework with compact scene representations from generataive model.
	Omniirectional DSO: Direct Sparse Odometry with Fisheye Cameras Hidenobu Matsuki, Lukas von Stumberg, Vladyslav Usenko, Jörg Stückler, Daniel Cremers IEEE Robotics and Automation Letters (RA-L), 2018 paper / video Real-time direct visual odometry for omnidirectional cameras

Academic Service

Conference Reviewer: NeurIPS, ICRA, IROS
Journal Reviewer: T-PAMI, IJCV, T-RO, IJRR, RA-L, ISPRS

Misc

I often use a photo of deer taken in Nara, Japan for my profile, as it's where I grew up and I also respect the groundbreaking PTAM demo connected with this place. Outside of research and work, I enjoy playing rugby on the muddy pitches of England.

Thanks.