Edit3K: Universal Representation Learning for Video Editing Components

arXiv:2403.16048v1 Announce Type: new
Abstract: This paper focuses on understanding the predominant video creation pipeline, i.e., compositional video editing with six main types of editing components, including video effects, animation, transition, filter, sticker, and text. In contrast to existing visual representation learning of visual materials (i.e., images/videos), we aim to learn visual representations of editing actions/components that are generally applied on raw materials. We start by proposing the first large-scale dataset for editing components of video creation, which covers about …

abstract aim animation arxiv components contrast cs.cv editing effects filter images learn materials paper pipeline representation representation learning six sticker text transition type types understanding universal video video creation videos visual