摘要
卡口场景下的人脸检测是视频智能监控的关键技术.然而,由于不同的人脸数据集的样本分布之间存在差异,在现有公开数据集上训练得到的人脸检测模型在卡口场景下难以取得令人满意的效果.为了解决上述问题,构建了一个卡口场景下的人脸数据集,并提出了一种简单且有效的模型重训练方法.该重训练方法能在模型检测人脸时,自适应地选取新的训练样本进行模型的重训练.在卡口场景测试集上的实验结果表明,该重训练方法能明显降低聚合通道特征模型的平均漏检率.
Face detection,in videos of passengers going through station ticket barriers,is a fundamental step of the intelligent videosurveillance. However, since face data from different datasets follow different distributions, models trained on public-face benchmarksfail to obtain satisfying results in the scene of station ticket barriers.To solve this problem,we first construct our own face dataset inthis special scene, and then propose a simple and effective re training strategy.This strategy seladaptively selects new samples to re-train a new model when a model is detecting faces.Experiments on test set from the scene of ticket barriers show that this strategysignificantly reduces the log-average miss rate of aggregate channel feature model,demonstrating the effectiveness of our re-trainingapproach.
出处
《厦门大学学报(自然科学版)》
CAS
CSCD
北大核心
2017年第3期429-436,共8页
Journal of Xiamen University:Natural Science
基金
国家自然科学基金(61572409
61402386
61571188)
福建省2011协同创新中心项目(闽教科[2015]75号)
关键词
人脸检测
卡口场景
重训练
聚合通道特征模型
face detection
station ticket barriers
re-training
aggregate channel feature model