Systems and methods are disclosed for predicting one or more characteristics of a animal by applying computational methods to image(s) of the animal to generate one or more metrics indicative of the characteristics. Embodiments determine predictors of characteristics by creating a sample library of animals of a particular type, determining facial descriptor measurements for each animal, determining relationships between facial descriptor measurements and additional library data, and selecting predictors from these relationships. Other embodiments predict characteristics of animals not in the library and, optionally, categorize animals for particular discipline, training, management, care, etc. based on the characteristics. Other embodiments predict characteristics and determine strategies for group(s) of animals using predicted characteristics of individual animals. Embodiments are broadly applicable to domesticated animals including dogs, cats, cattle, oxen, llamas, sheep, goats, camels, geese, horses, chickens, turkeys, and pigs. Other embodiments predict certain characteristics of humans, including certain cognitive or developmental disorders.