A picture is worth a thousand words. While there is a huge variety of molecular descriptors that can be used to indirectly describe a chemical structure, the image of a molecule—represented either as a wireframe or ball-and-stick scheme—is a trustworthy manner to account for the atomic composition, connectivity, size, and shape of a chemical compound. Furthermore, other chemical properties may be encoded into an image since it is a collection of pixels, which are digital units that can be renumbered to indicate, for example, the electronegativity of a chemical element. In molecular modeling, particularly in quantitative structure-property relationships (QSPR), the multivariate image analysis (MIA) gives rise to valuable molecular descriptors since the changes in pixel coordinates of an image originated from the replacement of a substituent in a molecule with another explain the variance in the property related to the chemical structure. This chapter reports the historical development of MIA-QSPR and compares this method with other approaches based on molecular descriptors of several dimensionalities. In addition, the background of the technique, a step by step to build an MIA-QSPR model, applications, perspectives, and a case study are presented herein. In overall terms, MIA-QSPR appears as an alternative method to aid the design of novel drugs, agrochemicals, and materials in general, straightforwardly and expeditiously.