Pre-Publication V3. An generative AI process that uses existing video input(s) to generate modified video output(s).designed to use existing video as a primary input to produce a modified or transformed output, conditioning generation on source footage rather than text or static images.
Deliberation Summary:
Remove intent-based phrasing and describe the process by what it produces, with examples of secondary inputs moved to the standards work.
Pre-Publication V2. A generative AI process designed to that uses existing video as a primary input to produce a modified or transformed output, conditioning generation on source footage rather than text or static images. Applications include style transfer, character replacement, and environment transformation while preserving the motion or performance of the original. Distinct from text-to-video generation, it is particularly relevant to post-production and visual effects workflows.
Deliberation Summary:
(1) Reword text that reads as promotional, and use intent-based phrasing. (2) Cross-references to a text-to-video term that does not appear in the list, suggest either adding that entry or directing the reference to the related entry on generative video models.
Pre-Publication V1. A generative AI process that uses existing video as a primary input to produce a modified or transformed output, conditioning generation on source footage rather than text or static images. Applications include style transfer, character replacement, and environment transformation while preserving the motion or performance of the original. Distinct from text-to-video generation, it is particularly relevant to post-production and visual effects workflows.