Abstract: A system, method and computer program product for generating a virtual avatar representing a user. Sensor data is obtained from a plurality of sensors mounted to the user. The sensor data includes force sensor data from force sensors positioned underfoot of the user and IMU data from a plurality of inertial measurement units (IMUs) mounted at respective IMU locations on the user. An activity type is determined from the sensor data. An estimated full-body posture of the user is determined by inputting the IMU data, force sensor data, and activity type to a machine learning model trained to estimate full-body posture. The estimated full-body posture specifies a limb position and a limb orientation for each limb of the user. The virtual avatar is generated based on the estimated full-body posture and can be displayed. The virtual avatar can be generated without the use of motion capture sensors.

Abstract: A system, method and computer program product for generating a virtual avatar representing a user. Sensor data is obtained from a plurality of sensors mounted to the user. The sensor data includes force sensor data from force sensors positioned underfoot of the user and IMU data from a plurality of inertial measurement units (IMUs) mounted at respective IMU locations on the user. An activity type is determined from the sensor data. An estimated full-body posture of the user is determined by inputting the IMU data, force sensor data, and activity type to a machine learning model trained to estimate full-body posture. The estimated full-body posture specifies a limb position and a limb orientation for each limb of the user. The virtual avatar is generated based on the estimated full-body posture and can be displayed. The virtual avatar can be generated without the use of motion capture sensors.

Abstract: A system, method and computer program product for generating a virtual avatar representing a user. Sensor data is obtained from a plurality of sensors mounted to the user. The sensor data includes force sensor data from force sensors positioned underfoot of the user and IMU data from a plurality of inertial measurement units (IMUs) mounted at respective IMU locations on the user. An activity type is determined from the sensor data. An estimated full-body posture of the user is determined by inputting the IMU data, force sensor data, and activity type to a machine learning model trained to estimate full-body posture. The estimated full-body posture specifies a limb position and a limb orientation for each limb of the user. The virtual avatar is generated based on the estimated full-body posture and can be displayed. The virtual avatar can be generated without the use of motion capture sensors.

Abstract: An ergonomic seat for a cycle and a manufacturing method of said seat comprise two seating surfaces that are curved upward and forward to allow a user to be supported continuously in the precise anatomical regions situated under each ischium, under the upper dorsal part of the thighs as far as a ventro-lateral part of the thighs in the region of the acetabulums, and passing via each of the creases of the superficial dorso-lateral region of the body which join the thigh to the buttock.

Abstract: An ergonomic seat for a cycle and a manufacturing method of said seat comprise two seating surfaces that are curved upward and forward to allow a user to be supported continuously in the precise anatomical regions situated under each ischium, under the upper dorsal part of the thighs as far as a ventro-lateral part of the thighs in the region of the acetabulums, and passing via each of the creases of the superficial dorso-lateral region of the body which join the thigh to the buttock.