30/03/2018 RELEASE

“VisionPose” α version will be launched sequentially from April 2, 2018 (Monday)

【Add on April 3, 2018】
【Add on April 17, 2018】


“VisionPose” α version will be launched sequentially from April 2, 2018 (Monday)

~ High precision AI framework detection system without depth sensor ~


NEXT-SYSTEM Co., Ltd. decided to start distributing ‘VisionPose’ under development as a highly accurate AI framework detection system without using a depth sensor as an α version from April 2, 2018 (Monday, 2018) It was.

“VisionPose” is a system that can detect human skeleton and depth information with only WEB camera using deep learning without relying on Kinect or other camera with depth sensor.

Originally we were promoting development as an alternative as Kinect was discontinued due to the use of our evolutionary AR signage “Kinesys”, but let me release the demo under development As a result, I received many reactions. Therefore, we think that if you are a troublesome person by replacing the same Kinect, we are working on development to accommodate various uses.

Currently, while we are working on further improvement of accuracy and ease of use, we will start offering as alpha version for those who received inquiries.

If you are planning to use, please contact the following inquiries, we will correspond in sequence from April 2, 2018.

Since it is a product under development, I think that there may be some inconvenience, but sincerely we will respond so, thank you for your continued understanding.

■ Five features of “VisionPose” α version


  1. Overwhelmingly high accuracy. There is less bone shake than similar products.
    By using deep learning to estimate the skeleton, it is possible to measure more accurately than skeleton detection with a conventional sensor-equipped camera.
  2. It also supports depth measurement.
    Since it can extract skeleton information only with WEB camera, it does not depend on depth sensor.

    You can measure the depth using a stereo camera (2 WEB cameras).
  3. Multiple bone detection is possible without limitation of number of people.
    Processing speed is constant even as the number of people increases because we are collecting multiple people together in one process. Processing speed is higher than bone detection by multiple people than similar products.
  4. Excellent usability. Offer with Kinect like SDK.
    We will provide it with an interface close to Kinect’s SDK.
  5. Since it does not use infrared rays, it can be used outdoors.
    It is possible to use outdoors which was difficult in Kinect.


■ How to use

If you are considering using “VisionPose” alpha version please contact us from the following URL in the form.
We will correspond in sequence from April 2, 2018 (Monday).
* If there are many inquiries, there is a possibility that it will take time to respond. Please acknowledge it.

【↓ Add on 3rd April 2018】
Received a lot of your application than expected, we have gotten your time now in the corresponding.
I’m very sorry but I am pleased if you acknowledge it.
Also, for future reference, I am very pleased if you can describe the company name, purpose of use, purpose etc. within the range that you can teach.

【↓ Add on 17th April 2018】
I will inform you about the current situation.

Currently, we are providing feedback in response to requests received from companies that provided alpha versions in anticipation, and we take precedence over corporate response.
For individuals, we are considering providing Kinect in a more user friendly form as a package with a camera like Kinect, so I think if you get time for a while now.
We will be happy to hear your voice at our website or SNS etc. as soon as preparation is ready.
Thank you for your continued development as we continue to develop.

・Contact form