Konica Minolta’s AI-Based Human Posture Recognition Technology Incorporated in the Upgraded Version of the “LOVOT” Family Robot
Creating Stronger Affinity between User and Robot
May 30, 2022
Tokyo (May 30, 2022) – Konica Minolta, Inc. (Konica Minolta) announced that FORXAI Imaging AI, the company’s proprietary AI-based pose estimation technology, has been incorporated in the upgraded version of LOVOT, a family robot developed by GROOVE X, Inc.
The upgraded version of LOVOT is capable of approaching a user when the user crouches down to make eye contact with it. The pose estimation technology enables LOVOT to estimate human posture in real-time and move accordingly, making the user feel closer to LOVOT and a stronger affinity with it.
How the “FORXAI Imaging AI” Pose estimation technology Works
Using AI, Konica Minolta’s pose estimation technology can estimate human posture by automatically detecting a human present in an image and quickly tracking their skeleton. While the conventional LOVOT can recognize the presence of a human but not their posture or action, the pose estimation technology enables LOVOT to recognize both the presence and action of a human and move toward the person.
Building on Konica Minolta’s track record of developing office and medical equipment with faster image processing, this technology boasts high image processing speed combined with high recognition accuracy, as well as low power consumption and cost, thanks to edge computing.
How the “FORXAI Imaging AI” Pose estimation technology Can Be Used
The pose estimation technology to recognize the presence of a human and estimate their posture at the same time can be effectively used for various purposes, such as human behavior and sports analysis and communication between human and computer, like LOVOT. In addition, the technology is used in the following settings.
|HitomeQ Care Support service for care facilities|
|A sensor mounted on the ceiling detects a resident’s movements such as waking up, getting out of bed, falling from the bed, and stumbling, and sends a notification along with footage to the smartphones of care staff.|
|Runalytic running form improvement system|
|Designed for “citizen runners,” this system estimates running posture in real time, and analyzes and visualizes the running form to help improve it.|