EUVIMEDEUVIMED entdecken
Crossref

Leveraging Efficient Text-Prompt for Interactive Spectral Reconstruction

Xuedan Wang, Jiaojiao Li, Haitao Xu, Rui Song, Yunsong Li, Qian Du

IEEE Transactions on Circuits and Systems for Video Technology · 2026-11 · 36 (11) · 16515-16527

DOI: 10.1109/tcsvt.2026.3720449 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Circuits and Systems for Video Technology, 2026-11. Verfasst von Xuedan Wang und 5 weiteren Autor:innen. Die Arbeit nennt 49 Literaturangaben. Gefördert von: National Key Research and Development Program, Shaanxi Provincial Science and Technology Department, Xi'an Science and Technology Bureau.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Current spectral reconstruction (SR) methods predominantly employ empirically crafted deep architectures to recover hyperspectral images (HSIs) from low-dimensional observations. However, the limited spectral cues and the high correlation in low-dimensional embeddings make single-modality reconstruction fundamentally underconstrained, hindering the recovery of fine-scale spectral variations. To address this issue, we introduce a multi-modal SR framework that incorporates text-derived linguistic guidance encoding task-specific cues and spectral descriptors, which are efficiently processed without significant computational overhead into the reconstruction pipeline. Based on this idea, we develop the Progressive Text Interaction Reconstruction Network (PTIR). PTIR employs a hierarchical interaction paradigm to inject text-derived semantics, such as task-specific information, at multiple processing stages, effectively mitigating unimodal limitations. In particular, the dedicated Focus-Weighted Attention (FWA) synergizes with Kolmogorov-Arnold Networks(KAN) to dynamically modulate the focus scope of attention heads, refining the granularity of interactions between features at each level and accomplishing the modulation of features. Meanwhile, a novel diagonal Core-Correlation perception module (DCCP) is proposed to preserve the essential correlations while discarding redundancy via diagonal compression, enhancing fine-grained pattern recognition and yielding a compact, semantically enriched latent space. In addition, this kind of design significantly reduces computational complexity and memory footprint. Extensive experiments across multiple benchmarks demonstrate that the proposed PTIR consistently surpasses existing state-of-the-art methods.

Förderung

  • National Key Research and Development Program
  • Shaanxi Provincial Science and Technology Department
  • Xi'an Science and Technology Bureau
  • "Scientist + Engineer" team of the Qin Chuang Yuan in Shaanxi Province
  • Xi'an Science and Technology Program and the 'Leading the Charge' initiative for the industrialization of core technologies in key industria

Zitieren

APA

Wang, X., Li, J., Xu, H., Song, R., Li, Y., & Du, Q. (2026). Leveraging Efficient Text-Prompt for Interactive Spectral Reconstruction. IEEE Transactions on Circuits and Systems for Video Technology, 36(11), 16515-16527. https://doi.org/10.1109/tcsvt.2026.3720449

Vancouver

Wang X, Li J, Xu H, Song R, Li Y, Du Q. Leveraging Efficient Text-Prompt for Interactive Spectral Reconstruction. IEEE Transactions on Circuits and Systems for Video Technology. 2026;36(11):16515-16527. doi:10.1109/tcsvt.2026.3720449

BibTeX
@article{Wang2026,
  author = {Wang, Xuedan and Li, Jiaojiao and Xu, Haitao and Song, Rui and Li, Yunsong and Du, Qian},
  title = {Leveraging Efficient Text-Prompt for Interactive Spectral Reconstruction},
  journal = {IEEE Transactions on Circuits and Systems for Video Technology},
  year = {2026},
  volume = {36},
  number = {11},
  pages = {16515--16527},
  doi = {10.1109/tcsvt.2026.3720449}
}
Kostenlos starten →