Video object segmentation based on motion-aware ROI prediction and adaptive reference updating

作者：

Highlights：

• This method achieves the speed-accuracy trade-off.

• The paper predicts the region of interest by perceiving the motion trend of target.

• The influence of the background is reduced by feeding the target ROI.

• The appearance changes of target are adapted by updating the reference dynamically.

• Temporal stability is improved by the adaptive updating strategy of reference.

摘要

•This method achieves the speed-accuracy trade-off.•The paper predicts the region of interest by perceiving the motion trend of target.•The influence of the background is reduced by feeding the target ROI.•The appearance changes of target are adapted by updating the reference dynamically.•Temporal stability is improved by the adaptive updating strategy of reference.

论文关键词：Video object segmentation,Region of interest prediction,Adaptive reference updating,Siamese network

论文评审过程：Received 27 July 2019, Revised 6 August 2020, Accepted 23 October 2020, Available online 29 October 2020, Version of Record 10 February 2021.

论文官网地址：https://doi.org/10.1016/j.eswa.2020.114153