A novel pixel neighborhood differential statistic feature for pedestrian and face detection

作者：

Highlights：

• Our proposed methods can be viewed as a simplified one-stage CNN, which can also achieve state-of-art results.

• We proposed four different types of pixel neighborhood differential features, which aim to mine the discriminative information of the intrinsic structure for the pedestrian.

• We proposed an unsupervised feature pattern learning method, which can reduce the redundancy of the feature and discover discriminative differential statistic patterns.

• We proposed a supervised feature pattern learning method, which is utilized to get more compacted and informative feature for pedestrian detection.

摘要

•Our proposed methods can be viewed as a simplified one-stage CNN, which can also achieve state-of-art results.•We proposed four different types of pixel neighborhood differential features, which aim to mine the discriminative information of the intrinsic structure for the pedestrian.•We proposed an unsupervised feature pattern learning method, which can reduce the redundancy of the feature and discover discriminative differential statistic patterns.•We proposed a supervised feature pattern learning method, which is utilized to get more compacted and informative feature for pedestrian detection.

论文关键词：Pedestrian detection,Face detection,PDF,Neighborhood differential statistic patterns

论文评审过程：Received 14 January 2016, Revised 21 July 2016, Accepted 18 September 2016, Available online 22 September 2016, Version of Record 6 October 2016.

论文官网地址：https://doi.org/10.1016/j.patcog.2016.09.010