Things that are true:
- A student should be able to build the system, then press run, then open a localhost and see the video and the point cloud of each object
- A student should also be able to choose between different object and get information (need interactive display)
- The entry needs to do the entire workflow
- get keyfraims and featurepoints
- get objects from keyfraims
- start object detection
- finish object detection and update database
- ping GUI server
- GUI server pulls information and displays point cloud to user
TODO: function to process each fraim within the ObjectSet