Video Golf Analysis Clubfitting Project

This project has recently changed directions from a more complex 3D inertial-type system to a more simplistic 2D video computer vision project, and time is of the essence to get it up and running competently. The attached document links to three short, two-dimensional videos of a golfer moving a golf club during what is termed the pre-swing period. The variations are fairly representative of what will be encountered in the real word. Certain pre-measured, known dimensions of the golf club can be input into an appropriate software analysis program and can include the precise total length of the club and the distances from its grip end to each of the florescent tape markers as needed. The general object is to track the club/markers as precisely as possible in the course of club movement and feed that data into an appropriate algorithm(s) to calculate a rotation point location occurring somewhere along the length of the club. The shown plane of movement is the critical one of interest, the one along which a sensation of effective clubhead weight (resulting from the rotation point location) is sensed by a golfer as a swing is about to begin. Different golfers can develop rotation points in substantially different locations along the length of a club, even when the exact same club is used to test with. A close examination can reveal that the club movement is not purely rotational in nature and that there can simultaneously be some generalized moving around of the club as a whole. This is typically the case among most golfers. But with that said, very small movement segments will commonly be isolated and analyzed, which might or might not limit any inaccuracies in results due to any overall (translational?) club movement. So the importance of accounting for this element is not firmly known yet. Much of the application can be performed manually. This includes a manual selection of frames viewed by the operator (just two potentially needed) within the specific segment of club movement that wants to be analyzed. There can also be a manual marking by the operator of the positions of the tracking markers on each frame (a feature that can actually be considered an advanced and not deficient feature of the application for analyzing various elements). This manual operation can substantially lessen the demands of elements like tracking marker type and intensity, lighting and background requirements, camera frame rate, and more. From there, computer-generated vectors can be developed and the proper math equations applied to the data as examples to calculate a rotation point location. I will need someone experienced in the applicable mathematics, physics, and/or computer vision techniques described to accurately make this application a reality as soon as possible. A simple desktop application will suffice at first for early analysis and testing of the application. But I do envision a web or phone application quickly following. In that light, I do not know whether the manual operation or interaction needed with the application as laid out above would prevent it from being usable as a web or phone application. I will need accurate recommendations regarding this before proceeding. I am certainly open to more automated tasks and automated computer-vision tracking of the club and markers, provided the requirements of achieving the tasks are not overly demanding and/or inconvenient for typical golfers (perhaps the largest market segment) to implement. This would also be important if such automated tasks would be essentially necessary in order for the application to function well as a web-based or phone application and not just desktop application. Any suggestions to help get at least a very basic version of this application up and running as soon as possible would of course be welcomed and appreciated. Thanks.