FSCM: Frequency-Enhanced Spatial-Spectral Coupled Mamba for Infrared Hyperspectral Image Colorization
The paper presents a new framework called FSCM for colorizing infrared hyperspectral images. This framework addresses limitations in existing methods by enhancing spatial-spectral coupling and improving texture details. Experimental results indicate that FSCM outperforms previous techniques in both visual quality and semantic fidelity.
- ▪FSCM is a spectral-information-guided GAN framework designed for infrared hyperspectral image colorization.
- ▪The framework includes a frequency-enhanced spatial-spectral state-space generator composed of cascaded FSB units.
- ▪FSCM improves semantic consistency in complex road scenes through an online semantic segmentation-guided loss.
Opening excerpt (first ~120 words) tap to expand
Computer Science > Computer Vision and Pattern Recognition arXiv:2605.15880 (cs) [Submitted on 13 May 2026] Title:FSCM: Frequency-Enhanced Spatial-Spectral Coupled Mamba for Infrared Hyperspectral Image Colorization Authors:Tingting Liu, Yuan Liu, Guiping Chen, Xiubao Sui, Qian Chen View a PDF of the paper titled FSCM: Frequency-Enhanced Spatial-Spectral Coupled Mamba for Infrared Hyperspectral Image Colorization, by Tingting Liu and 4 other authors View PDF HTML (experimental) Abstract:Thermal infrared imaging is robust to illumination variations and smoke interference, making it important for all-weather perception. However, the lack of natural color and fine texture limits target recognition, human visual interpretation, and the transfer of visible-light models.
…
Excerpt limited to ~120 words for fair-use compliance. The full article is at arXiv cs.AI.
Discussion
0 commentsMore from arXiv cs.AI
