Abstract: A system and method predicts one or more next steps during the construction or editing of a graphical model having executable semantics. The one or more next steps being valid actions according to the executable semantics and/or syntax of the graphical model. The system and method presents ghost versions of the one or more next steps on a display of the graphical model. In response to user selection of a given ghost version, the system and method changes the selected ghost version into a completed action at the graphical model. The system and system may also update an in-memory representation of the graphical model with the completed action. The in-memory representation supporting execution of the graphical model.

Abstract: A system for modifying video including a source of a filter stack specifying filter operations to be carried out on a video clip, a filter registry of filter functions including information on filter operations that can be carried out by the filter functions, and a filter planner that receives the filter stack and accesses the registry in order to generate a filter plan specifying filter functions to be used in carrying out the filter operations specified in the filter stack.

Abstract: A system for modifying video including a source of a filter stack specifying filter operations to be carried out on a video clip, a filter registry of filter functions including information on filter operations that can be carried out by the filter functions, and a filter planner that receives the filter stack and accesses the registry in order to generate a filter plan specifying filter functions to be used in carrying out the filter operations specified in the filter stack.