Visual saliency guided video compression algorithm
作者：
Highlights：
• A two-stage novel video compression architecture is proposed for video compression.
• Machine learning scheme over three dimensional features is used for saliency computation.
• Saliency computation at macroblock level saves computation.
• Thresholding of mutual information between successive frames indicates the frames requiring re-computation of saliency.
• The motion vectors propagate the saliency values for macroblocks in P frames.
摘要
Highlights•A two-stage novel video compression architecture is proposed for video compression.•Machine learning scheme over three dimensional features is used for saliency computation.•Saliency computation at macroblock level saves computation.•Thresholding of mutual information between successive frames indicates the frames requiring re-computation of saliency.•The motion vectors propagate the saliency values for macroblocks in P frames.
论文关键词：Visual saliency,Feature maps,Mutual information,QP tuning,Machine learning,Motion vector
论文评审过程：Received 14 August 2012, Revised 27 April 2013, Accepted 9 July 2013, Available online 16 July 2013.
论文官网地址：https://doi.org/10.1016/j.image.2013.07.003
