An image processing device includes an evaluation unit which acquires a search pattern which is an appearance pattern of a person in a scene of a video content containing the scene, where a similar scene is to be searched, and a search target pattern which is an appearance pattern of a person in a video content which is to be a search target of the similar scene, calculates a degree of similarity between the search pattern and the search target pattern, and determines the similar scene on a basis of the calculated degree of similarity.