Preprint
Oct 2026
Moving Forward with Video Saliency: A New Dataset and Benchmark where Motion Matters
SalTempto is introduced, a video saliency benchmark with greater dynamism: 224 clips of highly dynamic content, sourced from the HACS-Segments dataset so that each clip contains an event together with its lead-up and aftermath, with gaze recordings from up to 16 subjects and a training split for adapting pretrained mod...
Susmit Agrawal, Rebecca Wanner, Juliane Verwiebe et al.
· 0 citations