Detecting skeletal information from images and videos without using markers — Pose Estimation AI Engine “VisionPose” enters the world market on November 4th.

Japanese company NEXT-SYSTEM CO., Ltd. is entering the global market, with their independently developed pose estimation AI engine “VisionPose”, which detects human skeletal information from camera images, still images and videos without the use of markers, starting November 4, 2021, via their newly opened e-commerce site.

Marketing Technology News: Sensibill Named a Top Growing Company by The Globe and Mail for the Second Consecutive Year

What is VisionPose?

VisionPose is a high-precision AI posture analysis engine that detects human skeleton information on camera images, still images and videos, without the need for markers or special equipment. As an all-around SDK package, it can be applied to various fields in development and research. With the “Standard” edition of VisionPose and the use of 2 cameras, high-precision skeleton detection in 3D coordinates can be performed, and by using the VisionPose “Single 3D” edition, skeleton detection in 3D coordinates with only one camera is possible. Sales outside of Japan will start with two products, VisionPose Standard Windows C# and C++, but the number of supported platforms will be expanded sequentially.

“I believe that VisionPose will be of great use in various fields of pose estimation and behavior analysis throughout the whole world.”— CEO Yoshio Fujita

VisionPose Functions and Merits

・　Skeleton detection in real-time
・　Detection of multiple people’s skeletons
・　Skeletal detection of up to 30 points in the human body
・　SDK without usage restrictions (with a similar interface as the Kinect V2 SDK*)
・　Detection of skeletal information of camera images, still images, videos
・　Sample program and tools included for direct use right after purchase (BodyAndColor, VP Analyzer)
・　30-day free trial before purchase

*KINECT is a registered trademark of Microsoft Corporation in the United States and other countries

Who is using VisionPose?

With a wide range of application fields, such as motion analysis in sports and fitness, workflow analysis and hazard detection in factories, safety surveillance in child- and nursing care, motion caption in entertainment and gaming, VisionPose is currently used by a total of 250 clients, including several major Japanese companies, and has sold over 300 units as of November 2021.

Representative examples are Toyota Motor Corporations robot “Welwalk WW-2000”, which utilizes VisionPose to assist rehabilitation of lower limbs that were paralyzed by strokes or other injuries, the “Avex Street Dance Exam” App of Avex Management Inc, which performs score evaluation of dance skills through video analysis with the help of VisionPose, and Panasonic i-PRO Sensing Solutions Co., Ltd. using VisionPose for human detection with the i-PRO Camera.

Marketing Technology News: MarTech Interview with Phil Gray, Chief Innovation Officer at Interactions LLC

Comment from NEXT-SYSTEM’s CEO Yoshio Fujita

“It has been 3 years since we started selling VisionPose inside of Japan, and since then it has grown as a pose estimation AI engine together with its user base. We have worked with edge-side built-ins such as AI cameras and AI chips, as well as cooperated with face recognition AI systems, which has again broadened its fields of application. Steadily growing as the main product of our AI business, we are now looking forward to making VisionPose available to a global audience. I believe that VisionPose will be of great use in various fields of pose estimation and behavior analysis throughout the whole world.”

Overview of VisionPose products supported outside of Japan

Product Name: VisionPose Standard Windows C#, C++
Development License: US$4,500 → 15%OFF
Distribution License: US$450 → 15%OFF

Product Breakdown:

・　VisionPose Software
・　Sample Program “BodyAndColor” for real-time skeleton visualization (source code included)
・　Video and Still Image Analysis Program “VP Analyzer”
・　VisionPose Manual (API Reference included)
・　Other included files