The number of cardiovascular imaging studies is growing exponentially, and so is the demand to improve the efficacy of the imaging workflow. Over the past decade, studies have demonstrated that machine learning (ML) holds promise to revolutionize cardiovascular research and clinical care. ML may improve several aspects of cardiovascular imaging, such as image acquisition, segmentation, image interpretation, diagnostics, therapy planning, and prognostication. In this review, we discuss the most promising applications of ML in cardiovascular imaging and also highlight the several challenges to its widespread implementation in clinical practice.