Stitch-a-Demo: Video Demonstrations from Multistep Descriptions

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Wu, Chi Hsuan, Ashutosh, Kumar, Grauman, Kristen
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911565802897408
author Wu, Chi Hsuan
Ashutosh, Kumar
Grauman, Kristen
author_facet Wu, Chi Hsuan
Ashutosh, Kumar
Grauman, Kristen
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2503_13821
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Stitch-a-Demo: Video Demonstrations from Multistep Descriptions
Wu, Chi Hsuan
Ashutosh, Kumar
Grauman, Kristen
Computer Vision and Pattern Recognition
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.
title Stitch-a-Demo: Video Demonstrations from Multistep Descriptions
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2503.13821