I am currently working for an artist who wants to realise a project that is centered around using vvvv to build a multi-channel live editing system for 4k footage shot in the ports of Beirut.
The current idea is that we will categorise footage in folders according to the objects and actions depicted. Then we will use SRTReader to read out lines of subtitles, and the sift node to check if certain keywords associated with said objects and actions are mentioned. If they are we will trigger a random video from a selection of videos with the corresponding content.
We are looking for someone who can build a simple proof of concept of the player and logic. If you are interested to hear more please email me with your day rate and experience at firstname.lastname@example.org