Vide-omics: A genomics-inspired paradigm for video analysis
作者：
Highlights：
• An overview of exploitation of bioinformatics approaches in computer vision.
• A novel video analysis paradigm, vide-omics, inspired by genomics principles.
• A vide-omics-based pipeline for foreground extraction from freely moving cameras.
• Robust performance largely independent from camera motion and scene
• State-of-the-art results in presence of varied and complex camera motions.
摘要
•An overview of exploitation of bioinformatics approaches in computer vision.•A novel video analysis paradigm, vide-omics, inspired by genomics principles.•A vide-omics-based pipeline for foreground extraction from freely moving cameras.•Robust performance largely independent from camera motion and scene•State-of-the-art results in presence of varied and complex camera motions.
论文关键词：Computer vision,Freely moving camera,Genomics,Foreground detection,Segmentation,Scanlines
论文评审过程：Received 3 November 2016, Revised 29 September 2017, Accepted 8 October 2017, Available online 10 October 2017, Version of Record 7 December 2017.
论文官网地址：https://doi.org/10.1016/j.cviu.2017.10.003
