Full Body Pose Estimation for Sports Analysis

Welcome to RidgeRun's Full Body Pose Estimation for Sport Analysis Project		Full Body Pose Estimation for Sports Analysis on NVIDIA ®Jetson™ About Camera Abstraction Camera Calibration AprilTag Camera Calibration Chessboard Camera Calibration 3D Pose Estimation Pose Representation Skeleton Calibration Temporal Filter Kinematic Fitting Pose Optimization Pose Alignment Pose Correction Performance Introduction Required Hardware How to Get the Code How to Install the Library How to Use the Library Getting Started Contact Us
Overview
Pose estimation has become one of the topics of interest in the computer vision field. It constitutes a great method of interaction between computational systems and the real world. It is widely used in tasks such as biomedical motion analysis, sports analysis, or even as a motion capture technology to animate virtual characters. At RidgeRun we decided to take part, and we started our own research project based on three-dimensional human pose estimation. We developed a multi-view human 3D pose estimation system, which is constituted by two stages. In the first stage, we perform an initial estimation using the data provided by the camera system. Then, on the second stage, the initial pose estimation gets processed to assure bio-mechanical plausibility and temporal consistency. With a multi-camera setup as simple as the one shown in the following image: Error creating thumbnail: Unable to save thumbnail to destination Render showing a possible hardware setup. You will be able to estimate the three-dimensional pose of the user. Reaching great results as shown in our next video demonstration. If you are interested in learning more, please take a look at our wiki. You will find technical documentation, tutorials, examples, and much more!
RidgeRun support
RidgeRun provides support for Embedded Linux development, specializing in the use of hardware accelerators in multimedia applications. RidgeRun's products take full advantage of the accelerators exposed to perform transformations on the video streams achieving great performance on complex processing.