English

Stitch-a-Demo: Video Demonstrations from Multistep Descriptions

Computer Vision and Pattern Recognition 2026-04-07 v3

Abstract

When obtaining visual illustrations from text descriptions, today's methods take a description with a single text context - a caption, or an action description - and retrieve or generate the matching visual context. However, prior work does not permit visual illustration of multistep descriptions, e.g. a cooking recipe or a gardening instruction manual, and simply handling each step description in isolation would result in an incoherent demonstration. We propose Stitch-a-Demo, a novel retrieval-based method to assemble a video demonstration from a multistep description. The resulting video contains clips, possibly from different sources, that accurately reflect all the step descriptions, while being visually coherent. We formulate a training pipeline that creates large-scale weakly supervised data containing diverse procedures and injects hard negatives that promote both correctness and coherence. Validated on in-the-wild instructional videos, Stitch-a-Demo achieves state-of-the-art performance, with gains up to 29% as well as dramatic wins in a human preference study.

Keywords

Cite

@article{arxiv.2503.13821,
  title  = {Stitch-a-Demo: Video Demonstrations from Multistep Descriptions},
  author = {Chi Hsuan Wu and Kumar Ashutosh and Kristen Grauman},
  journal= {arXiv preprint arXiv:2503.13821},
  year   = {2026}
}
R2 v1 2026-06-28T22:24:36.882Z