Each of video content items captured by cameras is associated with metadata describing a time stamp indicating a capturing time and spatial information indicating features of space relating to a camera at the capturing time. In the disclosed content reproduction method, at least one target video content item to be reproduced is selected from the video content items based on the time stamps and pieces of the spatial information described in the respective pieces of metadata, then a reproduction start position of the selected video content item is determined based on a time stamp described in metadata associated with the selected video content item, and the selected video content item is reproduced from the determined reproduction start position.