Weakly supervised object localization and segmentation in videos

作者：

Highlights：

• A method to localize and segment objects in weakly labeled videos is proposed.

• The method relies on object appearance model and temporal consistency constraint.

• Chain structured graphical model formulation is used to localize an object.

• The method does not require labeled training data.

• The method is fully automatic and does not require any user interaction.

摘要

•A method to localize and segment objects in weakly labeled videos is proposed.•The method relies on object appearance model and temporal consistency constraint.•Chain structured graphical model formulation is used to localize an object.•The method does not require labeled training data.•The method is fully automatic and does not require any user interaction.

论文关键词：Weakly supervised,Object localization

论文评审过程：Received 20 September 2015, Revised 4 June 2016, Accepted 22 August 2016, Available online 17 September 2016, Version of Record 28 September 2016.

论文官网地址：https://doi.org/10.1016/j.imavis.2016.08.015