According to one embodiment, a video analyzer includes a still image detector, a characteristic detector, and a boundary position detector. The still image detector detects a still image in video frames of content including video and audio data, and create a still image group of substantially the same still images. The characteristic detector includes at least one of a monochrome detector that detects a monochrome video frame in the video frames of the content and a silence detector that detects a silent portion in which the sound volume of the audio data is lower than a predetermined threshold. The boundary position detector detects a position where the type of the content changes based on information on at least one of the monochrome video frame and the silent portion contained in a segment where the still image group is present and a segment where no still image group is present.