EUVIMEDEUVIMED entdecken
Crossref

Plug-and-Play Tri-Branch Invertible Block for Image Rescaling

Jingwei Bao, Jinhua Hao, Pengcheng Xu, Ming Sun, Chao Zhou, Shuyuan Zhu

Proceedings of the AAAI Conference on Artificial Intelligence · 11.04.2025 · 39 (2) · 1826-1834

DOI: 10.1609/aaai.v39i2.32177 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 11.04.2025. Verfasst von Jingwei Bao und 5 weiteren Autor:innen. Die Arbeit wurde bisher 2-mal zitiert (Crossref).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

High-resolution (HR) images are commonly downscaled to low-resolution (LR) to reduce bandwidth, followed by upscaling to restore their original details. Recent advancements in image rescaling algorithms have employed invertible neural networks (INNs) to create a unified framework for downscaling and upscaling, ensuring a one-to-one mapping between LR and HR images. Traditional methods, utilizing dual-branch based vanilla invertible blocks, process high-frequency and low-frequency information separately, often relying on specific distributions to model high-frequency components. However, processing the low-frequency component directly in the RGB domain introduces channel redundancy, limiting the efficiency of image reconstruction. To address these challenges, we propose a plug-and-play tri-branch invertible block (T-InvBlocks) that decomposes the low- frequency branch into luminance (Y) and chrominance (CbCr) components, reducing redundancy and enhancing feature processing. Additionally, we adopt an all-zero mapping strategy for high-frequency components during upscaling, focusing essential rescaling information within the LR image. Our T-InvBlocks can be seamlessly integrated into existing rescaling models, improving performance in both general rescaling tasks and scenarios involving lossy compression. Extensive experiments confirm that our method advances the state of the art in HR image reconstruction.

Zitieren

APA

Bao, J., Hao, J., Xu, P., Sun, M., Zhou, C., & Zhu, S. (2025). Plug-and-Play Tri-Branch Invertible Block for Image Rescaling. Proceedings of the AAAI Conference on Artificial Intelligence, 39(2), 1826-1834. https://doi.org/10.1609/aaai.v39i2.32177

Vancouver

Bao J, Hao J, Xu P, Sun M, Zhou C, Zhu S. Plug-and-Play Tri-Branch Invertible Block for Image Rescaling. Proceedings of the AAAI Conference on Artificial Intelligence. 2025;39(2):1826-1834. doi:10.1609/aaai.v39i2.32177

BibTeX
@article{Bao2025,
  author = {Bao, Jingwei and Hao, Jinhua and Xu, Pengcheng and Sun, Ming and Zhou, Chao and Zhu, Shuyuan},
  title = {Plug-and-Play Tri-Branch Invertible Block for Image Rescaling},
  journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
  year = {2025},
  volume = {39},
  number = {2},
  pages = {1826--1834},
  doi = {10.1609/aaai.v39i2.32177}
}
Kostenlos starten →