EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

HIT-Net: A high-information-capacity dual-channel transformer network for video snapshot compressive imaging

Qiangyou Li, Jingxia Chen, Pengwei Zhang, Yanhua Xue, Jinshou Tian, Liwei Xin, Duan Luo, Luxia Xu, Chao Ji, Wei Zhao

AIP Advances · 01.05.2026 · 16 (5)

DOI: 10.1063/5.0316603 ↗

Kurz gefasst

Fachartikel in der Zeitschrift AIP Advances, online erschienen am 04.05.2026. Verfasst von Qiangyou Li und 9 weiteren Autor:innen. Die Arbeit nennt 41 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Key Laboratory of Neutron Science and Technology, National Natural Science Foundation of China, High Spatio-Temporal Resolution Diagnostic System for Two-Dimensional Velocity Field of Shock Wave.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Video snapshot compressive imaging is a computational imaging technique based on compressed sensing that acquires high-speed frames through a 2D compressed measurement, followed by an algorithmic reconstruction. Current deep learning-based reconstruction methods operate on single-channel measurements and fail to fully capture complex inter-element dependencies, resulting in detail loss and artifacts in high-precision reconstruction tasks. To address these issues, we propose a high-information-capacity dual-channel transformer network. The network first introduces a dual-channel fusion structure that increases effective input information, thereby reducing the equivalent compression ratio and improving reconstruction quality. We further enhance the transformer by replacing conventional positional encodings with a relative Gaussian weighting scheme and introducing an adaptive multi-head structure together with variable cross section residual connections, significantly strengthening the modeling of inter-element dependencies and high-fidelity detail recovery. Experiments on six standard benchmark datasets (Kobe, Runner, Aerial, Crash, Drop, and Traffic) demonstrate that the proposed method markedly outperforms all compared approaches and is the only one that consistently exceeds 30 dB peak signal-to-noise ratio across all datasets.

Förderung

  • National Key Laboratory of Neutron Science and Technology
  • National Natural Science Foundation of China
  • High Spatio-Temporal Resolution Diagnostic System for Two-Dimensional Velocity Field of Shock Wave
  • Photon Project of the Xian Institute of Optics and Precision Mechanics, Chinese Academy of Sciences
  • National Key Scientific Instrument and Equipment Development Projects of China

Zitieren

APA

Li, Q., Chen, J., Zhang, P., Xue, Y., Tian, J., Xin, L., Luo, D., Xu, L., Ji, C., & Zhao, W. (2026). HIT-Net: A high-information-capacity dual-channel transformer network for video snapshot compressive imaging. AIP Advances, 16(5). https://doi.org/10.1063/5.0316603

Vancouver

Li Q, Chen J, Zhang P, Xue Y, Tian J, Xin L, et al. HIT-Net: A high-information-capacity dual-channel transformer network for video snapshot compressive imaging. AIP Advances. 2026;16(5). doi:10.1063/5.0316603

BibTeX
@article{Li2026,
  author = {Li, Qiangyou and Chen, Jingxia and Zhang, Pengwei and Xue, Yanhua and Tian, Jinshou and Xin, Liwei and Luo, Duan and Xu, Luxia and Ji, Chao and Zhao, Wei},
  title = {HIT-Net: A high-information-capacity dual-channel transformer network for video snapshot compressive imaging},
  journal = {AIP Advances},
  year = {2026},
  volume = {16},
  number = {5},
  doi = {10.1063/5.0316603}
}
Kostenlos starten →