HyTAS: A Hyperspectral Image Transformer Architecture Search Benchmark and Analysis

TL;DR

HyTAS introduces the first benchmark for transformer architecture search tailored to hyperspectral imaging, evaluating multiple methods across datasets and analyzing factors influencing search performance to guide future research.

Contribution

It presents HyTAS, a comprehensive benchmark for transformer architecture search in hyperspectral imaging, including evaluations and analysis to advance the field.

Findings

Identified the best transformer architectures for HSI classification.

Provided extensive evaluation of 12 methods across 5 datasets.

Analyzed factors affecting transformer search performance.

Abstract

Hyperspectral Imaging (HSI) plays an increasingly critical role in precise vision tasks within remote sensing, capturing a wide spectrum of visual data. Transformer architectures have significantly enhanced HSI task performance, while advancements in Transformer Architecture Search (TAS) have improved model discovery. To harness these advancements for HSI classification, we make the following contributions: i) We propose HyTAS, the first benchmark on transformer architecture search for Hyperspectral imaging, ii) We comprehensively evaluate 12 different methods to identify the optimal transformer over 5 different datasets, iii) We perform an extensive factor analysis on the Hyperspectral transformer search performance, greatly motivating future research in this direction. All benchmark materials are available at HyTAS.