Context For movie viewers, the movie posters are one of the first impressions which humans use to get cues about the movie content and its genre. Humans can grasp the cues like color, expressions on the faces of actors etc to quickly determine the genre (horror, comedy, animation etc). It has been shown that color characteristics of an image like hues, saturation, brightness, contour etc. affect human emotions. A given situation arouses these emotions in humans. If humans are able to predict genre of a movie by a single glance at its poster, then we can assume that the color characteristics, local texture based features and structural cues of posters possess some characteristics which could be utilized in machine learning algorithms to predict its genre. Content The movie posters are obtained from IMDB website. The collected dataset contains IMDB Id, IMDB Link, Title, IMDB Score, Genre and link to download movie posters. Each Movie poster can belong to at least one genre and can have at most 3 genre labels assigned to it. As the dataset also includes the IMDB score, it would be really interesting to see if movie poster is related to rating. Acknowledgements The IMDB Id for movies were obtained from MovieLens. The IMDB Link, Title, IMDB Score, Genre and link to download movie posters were obtained from IMDB website. Inspiration Does color plays an important role in deciding the genre of the movie? Can raw image pixels contain enough information to predict genre from movie? Does number of faces in the poster say anything about the movie genre? What is the most frequent color used in horror movies? Which features are important to predict animated movie genre? If a movie belong to more than one genre, can we predict them all? Can we use movie posters only to predict movie rating?