ARank: Toward specific model pruning via advantage rank for multiple salient objects detection

作者：

Highlights：

• Propose a new concept, salient priority criterion (SPC)

• Propose a new model pruning method, advantage rank (ARank)

• Elaborate completely proof from theoretical and mathematical analysis

摘要

•Propose a new concept, salient priority criterion (SPC)•Propose a new model pruning method, advantage rank (ARank)•Elaborate completely proof from theoretical and mathematical analysis

论文关键词：Model compression,Model pruning,Salient object detection,Multiple salient object detection,Computer vision

论文评审过程：Received 17 February 2021, Revised 27 April 2021, Accepted 27 April 2021, Available online 30 April 2021, Version of Record 8 May 2021.

论文官网地址：https://doi.org/10.1016/j.imavis.2021.104192