Discovering salient objects from videos using spatiotemporal salient region detection
作者：
Highlights：
• A unified approach for spatiotemporal salient region detection is proposed.
• A novel temporal salient region detection approach is presented.
• Experimental study shows an improved performance of the proposed approaches.
• Faster variants of the proposed approaches also present outperforming results.
• Proof-of-concept prototype of the proposed framework is provided.
摘要
•A unified approach for spatiotemporal salient region detection is proposed.•A novel temporal salient region detection approach is presented.•Experimental study shows an improved performance of the proposed approaches.•Faster variants of the proposed approaches also present outperforming results.•Proof-of-concept prototype of the proposed framework is provided.
论文关键词：Salient region detection,Temporal saliency,Optical flow abstraction,Spatiotemporal saliency detection,Saliency map
论文评审过程：Received 4 October 2014, Revised 2 July 2015, Accepted 2 July 2015, Available online 16 July 2015, Version of Record 29 July 2015.
论文官网地址：https://doi.org/10.1016/j.image.2015.07.004
