Image processing apparatus, image processing method, and program

An image processing method comprising:

• displaying, on a display apparatus, a segment map comprising segment-representing images representing video segments obtained by dividing content into video segments, wherein the video segments comprise collections of frames of one or more temporally continuous frames;
• controlling reproduction of the video segments corresponding to the segment-representing images according to user input comprising a reproduction operation requesting reproduction with respect to the segment-representing images;
• and
• clustering frames of the content into a plurality of clusters;
• wherein:
• in the controlling of the reproduction of the video segment, in a case where, during the reproduction of a video segment of the content, a reproduction operation is detected with respect to a segment-representing image of an other video segment of the content, the other video segment, which corresponds to the segment-representing image with respect to which the reproduction operation is detected, is reproduced while maintaining the reproduction of the video segment being reproduced;
• and
• the method further comprises:
• determining a maximum likelihood state series in which characteristic amounts in chronological series of frames of the content are observed using a state transition model having states and state transitions, and
• clustering each frame of the content into clusters corresponding to states in which characteristic amounts of the frames are observed in the maximum likelihood state series.

At least one non-transitory computer-readable medium storing computer-executable instructions that, when executed by at least one processor, cause a computer to execute a method comprising:

• displaying, on a display apparatus, a segment map comprising segment-representing images representing video segments obtained by dividing content into video segments, wherein the video segments comprise collections of frames of one or more temporally continuous frames;
• and
• controlling reproduction of the video segments corresponding to the segment-representing images according to user input comprising a reproduction operation requesting reproduction with respect to the segment-representing images;
• and
• clustering frames of the content into a plurality of clusters;
• wherein:
• in a case where, during reproduction of a video segment of the content, a reproduction operation is detected with respect to a segment-representing image of an other video segment of the content, the other video segment, which corresponds to the segment-representing image with respect to which the reproduction operation is detected, is reproduced while maintaining the reproduction of the video segment being reproduced;
• and
• the method further comprises:
• determining a maximum likelihood state series in which characteristic amounts in chronological series of frames of the content are observed using a state transition model having states and state transitions, and
• clustering each frame of the content into clusters corresponding to states in which characteristic amounts of the frames are observed in the maximum likelihood state series.