Refining touch-to-track Interaction for Camera Motion Control on Mobile Devices

Controlling a film camera to follow an actor or object in an aesthetically pleasing way is a highly complex task, which takes professionals years to master. It entails several sub-tasks, namely (1) selecting or identifying and (2) tracking the object of interest, (3) specifying the intended location in the frame (e.g., at 1/3 or 2/3 horizontally) and (4) timing all necessary camera motions such that they appear smooth in the resulting footage. Traditionally, camera operators just controlled the camera directly or remotely and practiced their motions in several repeated takes until the result met their own quality criteria. Automated motion control systems today assist with the timing and tracking sub-tasks, but leave the other two to the camera operator using input methods such as touch-to-track, which still present challenges in timing and coordination. We designed a refined input method called TrackLine which decouples target and location selection and adds further automation with even improved control. In a first user study controlling a virtual camera, we compared TrackLine to touch-to-track and traditional joystick control and found that the results were objectively both more accurate and more easily achieved, which was also confirmed by the subjective ratings of our participants.

In Proceedings of The 16th IFIP TC.13 International Conference on Human-Computer Interaction (INTERACT ’17) Mumbai, India