Handle User Center Labeled Images

[PotatoPalooza]

This will likely best be accomplished by an async function in the web server that handles the incoming images with labeled centers. Once COLMAP has ran and the image centers for ~6 images have been located the center of the object in the scene can be found by projecting rays from the center pixel in each image and finding the closest point to these 6 lines. With so few lines linear regression should be fine or the RANSAC algorithm could be used here. Once complete a new center offset of the scene should be stored in Mongodb.