HYDRA:面向高通道高光谱相机应用的混合知识蒸馏与光谱重建算法
HYDRA: HYbrid knowledge Distillation and spectral Reconstruction Algorithm for high channel hyperspectral camera applications
AI总结:
提出 HYDRA 混合知识蒸馏与光谱重建架构,通过 Teacher-Student 模型从三通道彩色图像恢复数百通道 HSI,在所有指标上达到 SOTA,准确率提升 18% 且推理更快。
摘要:
高光谱图像(HSI)有望支持计算机视觉中的一系列新应用。近期研究探索了可泛化光谱重建(SR)的可行性,即从自然三通道彩色图像中恢复 HSI,并将其应用于未见过的场景。然而,以往的多尺度注意力(MSA)研究仅在非常稀疏的光谱上证明了足够的可泛化结果,而现代 HSI 传感器包含数百个通道。本文通过我们提出的混合知识蒸馏与光谱重建架构(HYDRA),介绍了一种新的光谱重建方法。该方法使用一个封装潜在高光谱图像数据的 Teacher 模型,以及一个学习从自然图像到 Teacher 编码域映射的 Student 模型,并结合一种新的训练方法,实现了高质量光谱重建。这解决了以往 SR 模型的关键局限,在所有指标上均取得 SOTA 性能,包括准确率提升 18%,并且在不同通道深度下比当前 SOTA 模型具有更快的推理时间。
英文摘要:
Hyperspectral images (HSI) promise to support a range of new applications in computer vision. Recent research has explored the feasibility of generalizable Spectral Reconstruction (SR), the problem of recovering a HSI from a natural three-channel color image in unseen scenarios. However, previous Multi-Scale Attention (MSA) works have only demonstrated sufficient generalizable results for very sparse spectra, while modern HSI sensors contain hundreds of channels. This paper introduces a novel approach to spectral reconstruction via our HYbrid knowledge Distillation and spectral Reconstruction Architecture (HYDRA). Using a Teacher model that encapsulates latent hyperspectral image data and a Student model that learns mappings from natural images to the Teacher's encoded domain, alongside a novel training method, we achieve high-quality spectral reconstruction. This addresses key limitations of prior SR models, providing SOTA performance across all metrics, including an 18\% boost in accuracy, and faster inference times than current SOTA models at various channel depths.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。