Pose Estimation for Indoor Rowing Performance Analytics
Anton Ukhlin
Moscow
- 0 Collaborators
The main goal of this project is to describe an image processing approach capable of estimating the pose of athletes exercising on indoor rowing machines. Analyzing the image content may help improve the motion patterns of beginner rowing athletes at home workouts during the COVID-19 quarantine ...learn more
Project status: Concept
Mobile, Internet of Things, Artificial Intelligence
Intel Technologies
OpenVINO,
Intel Python
Overview / Usage
Rowing is a simple sport with complicated aspects. The complication is in that rowing includes elements of physiology, psychology, biomechanics, boat setup, teamwork, and an x-factor that we cannot define. Rowing is a sport that is purely about getting a boat from start to finish as fast as possible.
On-water rowing is available most times depending on location during the season (from mid-Spring to mid-Autumn), so indoor rower was designed for water athletes to stay fit on the land all year. However, the rowing technique is essential for each rowing athlete.
According to British Rowing, 1 of 2 indoor rowers are self-taught. Due to COVID-19 personal trainers, gym staff are hardly or not accessible for instant review of rowing technique most times. So there is an idea to use a smartphone camera for pose estimation during indoor rowing. And provide immediate feedback on how to improve rowing technique and how to use the rowing machines correctly.
Further project description in progress.
GLossary
An indoor rower is a machine used to simulate the action of watercraft rowing for the purpose of exercise or training for rowing. The indoor rower is calibrated to measure the amount of energy the rower is using through their use of the equipment.
Methodology / Approach
In progress
Technologies Used
In progress