Skip to main content
Keigo Minamida Profile Picture

I explore the expansion of human freedom in meaning-generation by supporting the inherent human editorial acts of "segmentation, relationalization, structuralization, and representation" at the level of media and tools in contemporary society where information technology (IT) digitizes all subjects. In our highly informatized society, a future is imminent where everything is captured as data and reconstructed as digital twins in virtual space. This digitalization trend is certainly heading toward a definitive endpoint. However, to derive value from informatized data, human editorial intervention—reinterpretation and reconstitution of information—is indispensable. My life theme lies in "editing as software" that goes beyond mere information collection and generation, enabling humans to freely and flexibly re-edit information. From an HCI (Human-Computer Interaction) perspective, I develop technical methods and interface designs that allow humans to freely re-edit information, proposing a new practical learning approach of "reading, writing, AI" in high-speed, high-volume information environments. This is an attempt to connect the information editing practices that humanity has historically accumulated to the present as "information cultural technology."

Research

Research Projects

Can People Distinguish Human and AI Agency in Humanoid Teleoperation?

UIST Adjunct '26

Can People Distinguish Human and AI Agency in Humanoid Teleoperation?

Can people distinguish human from AI in humanoid teleoperation?

Warping the Workspace

UIST Adjunct '26

Warping the Workspace

Teleoperation in a warped space.

Spatial Media Conference 2026

SCOPE-GS

Online sequential updates of 3D Gaussian Splatting in dynamic environments.

Augmented Leap

Augmented Humans 2026

Augmented Leap

Augmenting human jumps with thrust

PaperDemo (coming soon)
Incremental Gaussian Splatting

SIGGRAPH Asia 2024

Incremental Gaussian Splatting

System for gradual 3D reconstruction from monocular camera adapting to real-world changes

Recertif

SI2023

Recertif

System that enables checking robot work status by simply looking towards the robot

FSTL - Forest Segmentation & Trimming Learning

JSAI 2023

FSTL - Forest Segmentation & Trimming Learning

Improvements to pruning techniques (Slide created by Fujisaki)