Hyperspectral reflectance imaging in the short-wave infrared range (SWIR, “extended NIR”, ca. 1000 to 2500 nm) has proven to provide enhanced characterization of paint materials. However, the interpretation of the results remains challenging due to the intrinsic complexity of the SWIR spectra, presenting both broad and narrow absorption features with possible overlaps. To cope with the high dimensionality and spectral complexity of the datasets acquired in the SWIR domain, one data treatment approach inspired by innovative development in the cultural heritage field has been implemented: the use of a pigment spectral database (extracted from model and historical samples) combined with a deep neural network (DNN). This approach allows for multi-label pigment classification within each pixel of the data cube. The DNN model proposed in the study offers new possibilities in the SWIR range to identify and map pigments in complex materials either for unknown mixtures or multilayered systems. Conventional Spectral Angle Mapping (SAM) and DNN results obtained on both pigment reference samples and a Buddhist painting (thangka) will be compared. The DNN method, based on multiple inputs from mockup and historical paintings used as a training dataset, allows multi-labeling of the different zones and thus a more representative mapping of the pigments of interest.