MARs: Multi-Scale Convolution-Attention residual Fusion for Video Summarization.

Joon-Seok Song, Juyeob Lee, Eunil Park

Browse the full MMSys paper archive.