Pattern mining based video saliency detection
Hiba Ramadan, Hamid Tairi · 2017
A new spatiotemporal saliency detection model is presented in this paper. Instead of previous works which combine the image saliency in the spatial domain with motion cues to build their video saliency model, we propose to apply the pattern mining algorithm. From initial saliency maps computed in spatial and temporal domains, discriminative saliency patterns can be recognized and used to detect pertinent background and foreground seeds, then their label information is propagated to obtain the final saliency map. Our model ensures a good compromise between image saliency and motion saliency and presents an accurate prediction to estimate salient regions in comparison with other methods for video saliency detection.