Made in Japan

AI Behavior Analysis SystemVP-Motion

  • Employee Tracking
  • Patient Monitoring
  • Posture Correction

Real-time Behavior Analysis through Machine Learning
The AI Solution to Improve Efficiency in the Medical, Manufacturing and Sports Industries


What is VP-Motion?

Imagine the following. You own a factory and want to monitor your employees to see whether they are performing their duties correctly and safely. Or perhaps you are in charge of a medical facility and you want to keep an eye on your patients or clients. What if you are involved in sports, and you want to track one player in particular? These examples require a watchful eye and unwavering focus. But what if your eye strays, or you lose concentration? A potential disaster.

Fear no longer. VP-Motion is Japan’s leading AI-powered human behavior analysis software, that learns and automatically detects human movements. Through high-speed, video-based machine learning, the integrated AI automatically detects a wide range of movement and behavioral patterns. VP-Motion allows real-time detection of a diverse range of abnormal behaviors, from mistakes or accidents at the workplace—such as falls or people collapsing—to suspicious activities. VP-Motion is essentially an autonomous surveillance system with many functions.

Think of an emergency medical alert system. Or use it in sports. For example as a substitute AI fitness coach helping you with posture correction, or during sports rehabilitation. Whenever the app detects something strange, you get notified instantly. This reduces the need for continuous monitoring, optimizes workflow, prevents injuries, eliminates work errors and elevates worker safety. All in one small but powerful package.

New features added! Ver 1.2.3 update information

Adding the new video-based
analyzer ‘Analyzer’
・Calculation of time spent performing behavior and .csv output
・Select folder to automatically analyze all videos within ↓ Read more
New addition and improved
accuracy added to image-based analysis
New hand-crop function focuses just on hands
without interference from backgrounds or clothing ↓ Read more
Skeleton-based analysis gets
updated to latest pose estimation AI
・Improved accuracy and less interference from the background
・Freedom of camera positioning improved ↓ Read more

Tailor the System to Your Needs
Create Your Own Training Data

VP-Motion can be used in a wide variety of settings, and the user can freely customize the software to target specific behaviors. Here are just some examples of the multitude of possible fields and scenarios in which VP-Motion could be an asset.

Teach VP-Motion to detect routine tasks Workflow Optimization in Manufacturing

Teach VP-Motion to detect someone collapsing Medical Alert System

Teach VP-Motion to detect optimal movement patterns Sports Rehabilitation

Teach VP-Motion to detect people falling Patient and Elderly Care Support

Customer Reviews

A selection of reviews we have received from our many customers since first releasing VP-Motion in 2022.

Major construction company
We are able to supply the system with additional learning because VP-Motion comes with an annotator and learning function. At other companies, we would have to pay extra every time we would request additional learning.
Major car manufacturer
Because VP-Motion is a standalone program, we had no worries that our sensitive data would leak. Other companies use cloud computing, so we would have to upload our data to the cloud, creating unnecessary risks.
Major machineries manufacturer
We can perform real time analysis. Other companies only allow us to analyze prerecorded footage in batches.
Major parts manufacturer
Compared to other companies, VP-Motion has a much higher accuracy.
Multiple companies
The price is fantastic. At other companies, we would have to add at least one or two more zeroes to the final price.
Major car manufacturer
The fact that VP-Motion is a one-time purchase is great! Other companies only have subscription models.
Major car manufacturer
The fact that VP-Motion supports up to 8 cameras is perfect. Other companies limit us to just 1 camera.
Download PDF

System Components of VP-Motion

VP-Motion is comprised of three applications that all work together to let you generate trained AI-models through machine-learning, simply by feeding the program your own video data. The trained model is then used to monitor video images including, but not limited to, prerecorded video and live feeds.

System Benefits

Freely define behaviors

—– Easily create your data through the training data creation tool ‘Annotator’

A few videos and several minutes is all you need

—– High-speed machine-learning through state-of-the-art AI technology.

No more worrying about camera angles, VP-Motion detects everything

—– Thanks to our proprietary 3D-coordinate-detection technology, VP-Motion recognizes behaviors, even if
       the camera angle is different from the one used to train the model.

Annotate smarter and faster

—– Annotator can be freely copied to other PCs, allowing multiple people to annotate different videos
       at the same time, significantly cutting costs and boosting productivity.

Don’t be tied to one spot

—– ‘Standalone Monitor’ allows you to keep an eye on everything from whatever computer you choose
       outside of the installed environment, by simply loading the training data into the software.

No special equipment needed

—– VP-Motion supports all cameras, whether it’s a webcam or a surveillance camera;
       all cameras allow a fast detection of the skeleton.

Implement the system and start saving right away

As an example of what can be achieved by implementing VP-Motion specifically for manufacturing, we compared the man-hours and costs required to classify the content of video data for 100 employees performing their work manually versus using VP-Motion. Below, you will find an estimation of the potential savings.

Reduction of manhours

Save roughly 90% in terms of manhours

Reduction of costs

Save roughly 86% in terms of costs

How the System Works

Step 1Annotator: Label Behaviors to Create Training Data

The first step is to create training data. You do this by importing the video and labeling the desired behavior.

Annotator makes this process child’s play. All you have to do is draw a rectangle around the action and give it a name.
Click here to see a tutorial on how to use Annotator.

Step 2Trainer: Train the AI With Training Data

The second step involves teaching our AI-engine. You do this by importing the training data from step 1. This takes a couple of minutes, and a small amount of data.

After the importing is complete, and the engine has successfully learned the behavior you want it to, you will get a Trained AI Model.

Step 3Monitor: See everything with as much as eight cameras

Finally, in the Monitor software you will use the trained model you created in step two. You can use both real-time footage or prerecorded video. Monitor allows you to analyze the data through two different methods. These methods are ‘skeletal-based’ and ‘image-based’, and both methods allow you to detect and/or analyze behavior.

Merits of skeletal-based analysis

Skeletal-based analysis takes information from a person’s skeleton as its starting point. The system is most effective when there is a large amount of characteristics to detect in movement and actions, and it can detect these, even when just seeing the upper body of a person, or having multiple people overlap. Furthermore, the software can accurately learn someone’s movements in a short amount of time. Through this, checking routine procedures and human movements becomes a piece of cake. VP-Motion is useful when it comes to getting rid of slipups or accidents during work, because previously unnoticed details are suddenly clear as day. Thus, the application contributes greatly towards the increase in work effectiveness.

Ver1.2.3 brings support for latest pose estimation AI

Skeleton-based analysis now supports the latest in pose estimation AI, which brings improvements to the accuracy of its analysis, as well as reducing potential errors due to interference from the background.In particular, we have improved the freedom you get with regards to freely positioning cameras, and in doing so have made it even easier to smoothly integrate into your existing workspace. Finally, it’s now possible to switch between the existing and the new and improved AI.
Click here to see the video.

Merits of image-based analysis

Image-based analysis takes far more factors into consideration, such as background, objects and colors. It is most effective when there is a lack of pronounced movement, such as the detailed movements of a hand holding a tool, or the correct way to handle different tools, because this mode of detection can also detect factors outside of the human body, such as tools held in the hand or devices that are being worked on.

Classification changes depending on what is being held, such as a tool or machine component
Classification changes depending on the details of the work being performed, such as specific equipment being used.
Behavior analysis of images where only the arms or upper half of the body is being shown, such as work performed on a workbench.
Ver1.2.3 adds function to learn and analyze hands

The image-based analysis mode has had the ‘hand-crop’ function added to it, which allows VP-Motion to single out the hands, learn what they are, and then analyze them. Until now, VP-Motion would take all of the displayed pixels in the image into account, which caused irrelevant objects and backgrounds to interfere with the analysis. With the addition of the hand cropping mode, the surroundings will no longer be a factor, which allows you to create training data and analysis focused on the hands as well as being a huge increase to the accuracy of VP-Motion.
Click here to see the video.

Ver 1.2.3 allows export of time worked, as well as analysis of multiple videos

The newly introduced ‘VP-Motion Analyzer’ is an application meant for the analysis of prerecorded footage that allows you to calculate and export the time spent performing behavior which you have individually labeled. It also allows you to analyze multiple video files at once. Because you can now process them at high speed, the efficiency of analysis has taken a dramatic leap upwards.
Click here to see the video.

Analyzer’s merits
  • Analyze multiple videos in batches through folders, severely reducing required steps needed for analysis
  • No more time spent waiting, as real-time processing in the background speeds up your analysis
  • Combinable with every automatic system due to being operated through command line
  • Possibility to calculate time spent performing each video’s individually labelled behavior and export this to a .csv file, giving you the ability to use this as big data.

VP-Motion can also be used as surveillance camera recording software with the capability to log detected behavior. Users can also freely develop unique capabilities, such as sending out an alert when specific behavior is detected.

Listed below are some already accessible features
・Monitor and record up to eight cameras
・Export logged behavior analysis
・TCP/UDP communication through specified addresses and ports



VP-Motion All-in-One Package

VP-Motion for Windows
All-in-One Package

  • VP-Motion (Annotator/Trainer/Monitor) x 1
    Integrated environment for Annotator/Trainer/Monitor.
  • Standalone Monitor x 1
    Monitor app for a PC separate from the integrated environment.
  • Product Key for VP-Motion x 1
  • Product Key for Standalone Monitor x 1

PRICE $7,900$5,300

VP-Motion Standalone Monitor

VP-Motion for Windows
Standalone Monitor

  • Standalone Monitor x 1
    Monitoring app for a PC separate from the integrated environment.
  • Product Key for Standalone Monitor x 1

PRICE $990$675

Order Information

  • This product is provided as a download, and will not be physically shipped.
  • After your order is completed, we will email you the product key as well as a link to download the software in 1-3 business days. For any questions about our products, please contact us via this form.
  • Payments are handled by MyCommerce, an ecommerce platform by Digital River. For all matters related to purchase and payment, please see the MyCommerce FAQ, or contact MyCommerce directly.


Supported OSWindows 10 (64bit) / Windows 11
PC Specifications CPU: Intel® Core i7-6700 or equivalent, or higher
Memory: 32GB or more
Minimum required GPU : NVIDIA® GPU with a minimum of 6GB of VRAM
Recommended GPU: NVIDIA® GeForce RTX 3060 or equivalent*, or higher
*VP-Motion specifically requires a GPU with NVIDIA® chipset; NEXT-SYSTEM cannot guarantee the product’s functionality when an AMD GPU is used.
Number of Monitoring CamerasMax. 8 cameras per PC (Depending on PC performance)
Multiple monitoring PCs can be used through utilization of Standalone Monitor.
Download PDF

News / Topics

Release Notes

  • ver 1.2.3_EN
    06/28/2024 Major update includes:
    Added Analyzer: Analyzes multiple videos simultaneously and calculates time spent on individually labeled behaviors.
    Enhanced Image-based Analysis:Includes a new hand-crop function allowing VP-Motion to isolate and analyze hands and tools they hold.
    Updated Skeleton-based Analysis:Upgraded to the latest pose estimation AI for improved accuracy with reduced background interference.

    Also, we’ve added video playback features and addressed various bug fixes. Check out the manual for more info.
  • ver 1.1.6_EN
    10/02/2023 First Version