Hyperspectral imaging has been recognized as a powerful tool for characterizing complex marine benthic communities, yet its application to polar benthic ecosystems remains unexplored. In this study, we present the first field-based spectral library of living Antarctic benthic organisms, acquired during a field campaign on sea ice at Cape Evans, Ross Sea, combining two data collection platforms: a laboratory imaging system and HICYBOT, a vehicle customised for seafloor hyperspectral imaging (Cimoli et al. 2026). Samples of macroalgae, echinoderms, and substrates were collected by ROV and individually imaged in vivo under controlled lighting and water conditions, enabling sub-centimetre reflectance mapping across the visible to near-infrared spectrum (400–1000 nm). Reproducible reflectance correction and pixel segmentation workflows were applied to extract spectral signatures, which were then compared against community-scale seafloor imagery collected by HICYBOT. Reflectance spectra served as the common link bridging individual- and community-scale observations across both platforms. Normalized Difference Vegetation Index (NDVI) revealed photosynthetic activity in red macroalgae and epibiotic layers. Second derivative analysis identified diagnostic pigment absorption wavelengths at 570, 620, 655, and 700 nm, which were used to generate spectrometric pigment distribution maps at both scales, revealing phenotypic variation in pigment composition between growth forms and taxa. This study demonstrates a multi-scale methodological framework for Antarctic seafloor hyperspectral mapping, supported by the first spectral library of Antarctic benthic taxa. Our findings highlight the potential of underwater remote sensing platforms and human-led approaches to reveal hidden microspatial information in benthic communities, opening new research avenues for biodiversity monitoring and ecosystem processes.