HyperCon: Image-to-video model transfer for video-to-video translation tasks

Ryan Szeto, Mostafa El-Khamy, Jungwon Lee, Jason J. Corso

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

5 Scopus citations

Abstract

Video-to-video translation is more difficult than image-to-image translation due to the temporal consistency problem that, if unaddressed, leads to distracting flickering effects. Although video models designed from scratch produce temporally consistent results, training them to match the vast visual knowledge captured by image models requires an intractable number of videos. To combine the benefits of image and video models, we propose an image-to-video model transfer method called Hyperconsistency (HyperCon) that transforms any well-trained image model into a temporally consistent video model without fine-tuning. HyperCon works by translating a temporally interpolated video frame-wise and then aggregating over temporally localized windows on the interpolated video. It handles both masked and unmasked inputs, enabling support for even more video-to-video translation tasks than prior image-to-video model transfer techniques. We demonstrate HyperCon on video style transfer and inpainting, where it performs favorably compared to prior state-of-the-art methods without training on a single stylized or incomplete video. Our project website is available at ryanszeto.com/projects/hypercon.

Original languageEnglish
Title of host publicationProceedings - 2021 IEEE Winter Conference on Applications of Computer Vision, WACV 2021
Pages3079-3088
Number of pages10
ISBN (Electronic)9780738142661
DOIs
StatePublished - Jan 2021
Event2021 IEEE Winter Conference on Applications of Computer Vision, WACV 2021 - Virtual, Online, United States
Duration: 5 Jan 20219 Jan 2021

Publication series

NameProceedings - 2021 IEEE Winter Conference on Applications of Computer Vision, WACV 2021

Conference

Conference2021 IEEE Winter Conference on Applications of Computer Vision, WACV 2021
Country/TerritoryUnited States
CityVirtual, Online
Period5/01/219/01/21

Fingerprint

Dive into the research topics of 'HyperCon: Image-to-video model transfer for video-to-video translation tasks'. Together they form a unique fingerprint.

Cite this