What is 

Deep Runner?


01. What is Deep Runner?


The Deep Runner AI set-top box is an artificial intelligence-based general-purpose visual information recognizer that accurately recognizes the desired objects from the camera's high-resolution video and informs your service system.

AI experts and expensive AI systems are not necessarily required to implement new AI services. 

Your IT skills and a small Deep Runner AI set-top are enough.

02. Deep Runner Models

Deep Runner Z1

Deep Runner Z4



03. Deep Runner has the following advantages:



  • An expensive and large system is not required for AI services. One small device the size of the palm of your hand is enough. Because it consumes little power, it can also be operated with batteries.


  • AI experts are not required. When recognizing common objects such as people and cars, you can set it with just a few clicks through a simple web interface. Even if deep learning training is required for custom objects, it can be accomplished easily by following the guide provided. In addition, our technical staff will assist you if necessary. All the source codes used in our demo videos will be provided to the users.


  • Recognize images with the highest accuracy. Unlike most mobile AI devices that perform integer computations, Deep Runner computes deep learning algorithms using floating point computations, and little errors occur on the device.

(In this figure, 16-bit floating-point computation result completely overlaps 64-bit floating-point result, and is not visible on the graph.)

  • Smaller objects in the video can be recognized. Typical deep learning object recognition algorithms take a small image of about 300x300 pixels as an input, so there are limitations. Deep Runner supports high-resolution algorithms such as those with 600x600, 900x900 and 1080x1080 pixel inputs as well as 300x300 input algorithm, so a large amounts of information could be collected simultaneously from one camera.
  • By providing an independent computer environment within the device, service logics can be implemented within the device without a separate computer. Deep Runner runs on a latest Linux operating system and developers can use the entire system as a development platform.


04. Function


Deep runner supports Classification algorithm and Detection / Location algorithm. Select the algorithm according to the nature of the application you want.


  • Classification : Tells what the image is about.
  • Detection/Location : Tells where objects are.
  • Multiview : Up to 16 channels per Deep Runner can be recognized simultaneously.



05. What objects can be recognized?

  • It can recognize various categories such as gender, age, emotion, clothing as well as all objects with characteristic. (* Individual facial recognition is not supported at present.)
  • Deep Runner can also be recognized if they are visible and distant from the screen. However, it is possible to distinguish between size and length when camera zoom function input to Deep Runner and object position are fixed.



For more technical information, see the Deep Runner User Manual.