CoVEBench: Can Video Editing Models Handle Complex Instructions?
Researchers have introduced CoVEBench, a new benchmark designed to evaluate the capabilities of text-guided video editing models. This benchmark addresses the limitations of existing models that struggle with complex, multi-step editing instructions. CoVEBench comprises numerous videos and editing instructions, assessing models on their ability to comply with instructions and maintain video fidelity, revealing that current models often fail to perform multiple edits simultaneously or preserve content accurately. AI
IMPACT Highlights current limitations in AI video editing, pushing for development of models that can handle complex, multi-step instructions and preserve content.