AI technology is advancing to the point where it can recognize and analyze human behavior (motion).

Based on our own motion capture studio, we continuously acquire high-quality motion learning data, as well as image and video data. We utilize this data to conduct AI research and develop customized solutions in the motion field. to classification, detection, and segmentation, to extract meaningful information from multimedia data captured by various vision-based sensors, such as smartphones, cameras, CCTV, RGB-D, and LiDAR, analyze it, and derive desired results.

• 

  2D Pose estimation

  Technology that infers 2D joint coordinates of the human body in images.
  It is used in various fields such as sports analysis, motion recognition,
  nd AR/VR. It possesses the technology to accurately estimate postures even in unusual postures,
  such as Taekwondo movements, rather than everyday movements. It also possesses technology to
  generate data sets for special movements using a motion capture studio.
• 

  3D Pose estimation

  Estimate 3D human poses using one or multiple cameras.
  In addition to open datasets, you can create your own
  3D dataset using Motion Capture Studio to develop 3D pose estimation models.
  Since manual labeling of 3D data is extremely limited, o
  ur technology offers significant advantages in model development.

• 

  Action recognition

  Motion recognition can be broadly categorized into
  skeleton-based and image-based (RGB-based) action recognition.
  The appropriate action recognition model is selected based
  on the task, and our company possesses technology in both.
  By combining object recognition and action recognition,
  we can recognize the type of action of each object on a single screen.
  We hold one registered patent related to Taekwondo action recognition.