Although visual content prevails in the digital media environment, previous scholarship that attempts to detect bias and stereotypes in media content has mostly focused on textual data. Meanwhile, recent advances in computer vision have made the analysis of visual data on a large scale possible. Drawing theoretical insights from media bias, social cognition, visual persuasion, and gender studies literature, this dissertation investigates how various computer vision techniques—such as facial recognition, emotion detection, and computational aesthetics—can help us better analyze media bias in visual representations of politicians. In particular, study 1 examines partisan bias in media coverage of presidential candidates in the 2016 presidential election. Study 2 analyzes gender bias in politicians’ self-presentations on Instagram. In addition, this dissertation also hopes to illuminate the promises and caveats of using computer vision algorithms as data analysis tools.