An image capturing apparatus includes: an image input unit that inputs an image; a designating unit that receives designation of an initial tracking-target region that is a first region in a first image frame of time-series image frames input from the image input unit; a feature value extracting unit that extracts a predetermined feature value from a target region in the first image frame; a first search unit that searches for a second region obtained by changing a size of the first region and by determining whether the feature value extracted from the second region satisfies a predetermined condition for enabling successful tracking and sets the second region as a new initial tracking-target region; and a second search unit that searches for a region similar to the newly set initial tracking-target region as a tracking-result region in a second image frame subsequent to the first image frame.