Several different microscopy techniques optical, Optical-electron correlative microscopy (OECM), Raman and infrared (IR) spectroscopy, atomic force microscopy (AFM), X-ray computed tomography (X-ray CT), and focused ion beam-scanning electron microscopy (FIB-SEM) when used in combination reveal important and complementary details of shales at multiple scales and resolutions. OECM is effective at distinguishing variations in porosity distributions of shales and how they change as thermal maturity progresses, and relate these changes with specific organic matter type. AFM can provide high-resolution images at the nanoscale, providing insights in how shale pore structure, mineral distributions and micro-mechanical properties varies as thermal maturity progresses. AFM-IR is useful for assessing nano-scale geochemical and mechanical properties at much improved resolutions compared to IR used in isolation. X-ray CT images can discern pore-system features and assess shale anisotropy/heterogeneity at multiple scales. Deep-learning techniques, particularly various configuration of convolutional neural networks (CNN), are effective for CT image segmentation to distinguish features of interest and assess their distributions. Various super-resolution techniques are able to enhance the resolutions of low-resolution recorded CT images to discern very-small features. Applied to FIB-SEM images of various shale formations, the KiU-Net multi-branched configuration of CNN has shown its capability to resolve nanopores and micro-fractures in the size range 10–100 nm. These techniques offer the potential to substantially expand the level of detail extracted at multiple scales from shale images providing more effective formation characterization.