Elevating Flow-Guided Video Inpainting with Reference Generation
Proceedings of the AAAI Conference on Artificial Intelligence · 11.04.2025 · 39 (3) · 2527-2535
DOI: 10.1609/aaai.v39i3.32255 ↗Kurz gefasst
Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 11.04.2025. Verfasst von Suhwan Cho und 3 weiteren Autor:innen. Die Arbeit wurde bisher 7-mal zitiert (Crossref).
Abstract (Originalsprache)
Video inpainting (VI) is a challenging task that requires effective propagation of observable content across frames while simultaneously generating new content not present in the original video. In this study, we propose a robust and practical VI framework that leverages a large generative model for reference generation in combination with an advanced pixel propagation algorithm. Powered by a strong generative model, our method not only significantly enhances frame-level quality for object removal but also synthesizes new content in the missing areas based on user-provided text prompts. For pixel propagation, we introduce a one-shot pixel pulling method that effectively avoids error accumulation from repeated sampling while maintaining sub-pixel precision. To evaluate various VI methods in realistic scenarios, we also propose a high-quality VI benchmark, HQVI, comprising carefully generated videos using alpha matte composition. On public benchmarks and the HQVI dataset, our method demonstrates significantly higher visual quality and metric scores compared to existing solutions. Furthermore, it can process high-resolution videos exceeding 2K resolution with ease, underscoring its superiority for real-world applications.
Zitieren
APA
Cho, S., Oh, S. W., Lee, S., & Lee, J. (2025). Elevating Flow-Guided Video Inpainting with Reference Generation. Proceedings of the AAAI Conference on Artificial Intelligence, 39(3), 2527-2535. https://doi.org/10.1609/aaai.v39i3.32255
Vancouver
Cho S, Oh SW, Lee S, Lee J. Elevating Flow-Guided Video Inpainting with Reference Generation. Proceedings of the AAAI Conference on Artificial Intelligence. 2025;39(3):2527-2535. doi:10.1609/aaai.v39i3.32255
BibTeX
@article{Cho2025,
author = {Cho, Suhwan and Oh, Seoung Wug and Lee, Sangyoun and Lee, Joon-Young},
title = {Elevating Flow-Guided Video Inpainting with Reference Generation},
journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
year = {2025},
volume = {39},
number = {3},
pages = {2527--2535},
doi = {10.1609/aaai.v39i3.32255}
}
EUVIMED