Preprint
Aug 2026
Coverage-Driven Adaptive Keyframe Selection for Video Understanding
CSES is a training-free semantic keyframe selector that adaptively determines the numbers of frames to score and keyframes to select that formulates keyframe selection as a coverage problem that jointly accounts for semantic relevance, temporal redundancy, and visual redundancy.
Jun-Yang Zhang, Pu-Han Luo, Chenyu Tang et al.
· 1 citation