M3L: Language-based Video Editing via Multi-Modal Multi-Level Transformers
This addresses the difficulty for novices in using video editing tools by enabling automatic editing via natural language, though it is incremental as it builds on existing vision-and-language research.
The paper tackles the problem of making video editing more accessible by introducing a language-based video editing task, where a model edits a source video into a target video guided by text instructions, and proposes the M3L transformer that achieves effective results as shown in experiments on new datasets.
Video editing tools are widely used nowadays for digital design. Although the demand for these tools is high, the prior knowledge required makes it difficult for novices to get started. Systems that could follow natural language instructions to perform automatic editing would significantly improve accessibility. This paper introduces the language-based video editing (LBVE) task, which allows the model to edit, guided by text instruction, a source video into a target video. LBVE contains two features: 1) the scenario of the source video is preserved instead of generating a completely different video; 2) the semantic is presented differently in the target video, and all changes are controlled by the given instruction. We propose a Multi-Modal Multi-Level Transformer (M$^3$L) to carry out LBVE. M$^3$L dynamically learns the correspondence between video perception and language semantic at different levels, which benefits both the video understanding and video frame synthesis. We build three new datasets for evaluation, including two diagnostic and one from natural videos with human-labeled text. Extensive experimental results show that M$^3$L is effective for video editing and that LBVE can lead to a new field toward vision-and-language research.