Human Postures Estimation / Motion Recognition and Tracking

Using deep learning, we are developing an algorithm that recognizes and understands the three-dimensional posture and body shape of the human body from monocular color images. We are also developing a human body tracking algorithm that can track multiple people in the crowd at the same time. Aiming for a wide range of applications such as biomechanics, fashion, motion transfer to robots, and utilization in unmanned store systems.