Weakly supervised vehicle detection in satellite images via multi-instance discriminative learning

作者：

Highlights：

• A user-friendly, scheme is proposed to efficiently collect large-scale vehicle annotations.

• A density estimation algorithm with integer programming is proposed to learn from weak labels to estimate the initial vehicle location.

• A large-margin classification termed MIL-SVM is adopted to learn refined vehicle detector.

摘要

Highlights•A user-friendly, scheme is proposed to efficiently collect large-scale vehicle annotations.•A density estimation algorithm with integer programming is proposed to learn from weak labels to estimate the initial vehicle location.•A large-margin classification termed MIL-SVM is adopted to learn refined vehicle detector.

论文关键词：Multiple instance learning,Density estimation,Multiple instance SVM,Vehicle detection

论文评审过程：Received 1 June 2016, Revised 17 September 2016, Accepted 30 October 2016, Available online 3 December 2016, Version of Record 24 December 2016.

论文官网地址：https://doi.org/10.1016/j.patcog.2016.10.033