DISCRIMINATIVE TECHNIQUES FOR KEYFRAME SELECTION (FriPmPO1)

Author(s) : Matthew Cooper (FX Palo Alto Laboratory, United States of America) Jonathan Foote (FX Palo Alto Laboratory, United States of America)

Abstract : A convenient representation of a video segment is a single ``keyframe.'' Keyframes are widely used in applications such as non-linear browsing and video editing. With existing methods of keyframe selection, similar video segments will result in very similar keyframes, with the drawback that actual differences between the segments may be obscured. We present methods for keyframe selection that maximize two criteria: 1) capturing the similarity to the represented segment, so the keyframe is a good representation, and 2) preserving the differences from other segment keyframes, so that even different segments will have visually distinct representations. We present two discriminative keyframe selection methods, and an example of experimental results.

Menu