Answers are grounded in this paper and cite their sources.
Hyperspectral imaging (HSI) and augmented reality (AR) act as a symmetric pair of sensing operations: HSI encodes a physical object into a spectral response at the acquisition end, and AR decodes that response back into physical space at the display end. Registration and calibration keep the two ends consistent, so that the composition of the two maps stays close to the identity. This review organizes HSI-AR fusion as a closed loop of acquisition, multimodal registration, spatial anchoring and augmented visualization, and reads it in symmetry terms: rigid-body symmetry between sensor and display frames, symmetry breaking of an otherwise flat spectrum as the source of material contrast, asymmetric pairing of spectral and spatial sensors, and the mismatch between symmetric technical metrics and asymmetric task-error costs. Applications in surgery, mineral analysis, education and cultural heritage are compared by architecture, with cultural heritage treated as the leading case. Systems divide by whether spectra are acquired in real time or beforehand. The main challenges are spatiotemporal consistency, end-to-end latency, task-oriented representation of high-dimensional spectra, and cross-system evaluation. Future work will center on acquisition–computation co-optimization, multisensor fusion, edge computing and task-oriented interaction. The framework offers a common reference for HSI-AR systems in the field.
No takes yet. Share an insight, caveat, or question.
Xu et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: