Capturing video clips is often thought to be the first step in the video creation process. Of course it is not as a lot of preparatory effort has to be put into planning a shoot if professional results are expected. Video technology advances and people can more and more often just push rec on their mobile phone on the spurr of the moment. Nevertheless even in such an instantaneous decisions selecting the capture format (still/moving image/audio) has to be made.
The best way to capture the moment depends on the artist’s personal preferences, but a lot can be also reasoned if the end results and the environment are known. Due to its self-explanatory nature capturing activity would have been a good starting point for discussing automated video systems. For the sake of coherence we will start with the first part, preparation.
We have established that capturing is seldom the first step and this is the case even more so for the purposes of automated video systems. In a traditional production a lot of work goes on before the rec-button is pressed or the firsts ray of light hits frame #1 of film. The cameras must be installed on location beforehand with the best camera angles. Script for the impending call to “action” has to be written and rehersed in advance. Camera settings have to be adjusted to light conditions and wanted look. Only after all this is set the shoot can actually begin.
Intelligent capturing where triggers control the process are currently possible only to scenarios limited to strictly confined situations. Current applications include security and surveillance, where the defensive “script” describes the environment to be protected. The key areas are marked with motion triggers and admission of the subjects in the scene environment are controlled by other sensors. In a strict sense of automated video systems every recording surveillance camera on this world is capturing the “play of life” from every day of the year.
Advanced security and military technology is often reused for entertainment purposes. All technologies become common-place only if people accept the tradeoff between consequences and added value generated by using the technology. In the not so distant future a pick-up game of basketball maybe recorded by the friendly neighborhood surveillance camera every afternoon from 6-8pm. Further away lies a world where much more complicated rules on are given to a mobile system before unleaching it to the chosen environment. An interesting balance has to be found and set of rules will have to be met for such systems to exist.
Planning in such a complicated environment is going to need tools for modeling the scene. Adhering to rules will be easier using templates generated from successful applications. An experienced operator might be able to come up with such models. However experience comes only from learning through diverse projects. In the lond run the operater can hopefully be omited all-together with re-usable models or design patterns.