4

基于深度图的人体检测 HumanDetectionUsingDepth

 2 years ago
source link: https://www.oschina.net/p/humandetectionusingdepth
Go to the source link to view the article. You can view the picture content, updated content and better typesetting reading experience. If the link is broken, please click the button below to view the snapshot at that time.
neoserver,ios ssh client

该项目是一系列用深度图进行人体检测的一些机器学习方法的实现,主要方法是通过传统特征算子提取特征,然后训练SVM分类器,通过多尺度滑窗的方式处理整幅图像。可用于基于机器学习进行深度图人体检测方法的学习或项目。

包括的论文

HDD:
Wu S, Yu S, Chen W. An attempt to pedestrian detection in depth images[C]//Intelligent Visual Surveillance (IVS), 2011 Third Chinese Conference on. IEEE, 2011: 97-100.
SLTP:
Yu S, Wu S, Wang L. Sltp: A fast descriptor for people detection in depth images[C]//Advanced Video and Signal-Based Surveillance (AVSS), 2012 IEEE Ninth International Conference on. IEEE, 2012: 43-47.
ELDP:
Zhong F, Zhang J. Face recognition with enhanced local directional patterns[J]. Neurocomputing, 2013, 119: 375-384.
LDP: Jabid T, Kabir M H, Chae O. Facial expression recognition using local directional pattern (LDP)[C]//2010 IEEE International Conference on Image Processing. IEEE, 2010: 1605-1608.
LTDP:
Shen Y, Hao Z, Wang P, et al. A novel human detection approach based on depth map via Kinect[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops. 2013: 535-541.

项目依赖OpenCV3, 假设OpenCV位于D://opencv, 则将CMakeLists.txt中OpenCV_DIR设置为此路径,以方便找到OpenCV.
项目目前支持Windows,但只是一些目录操作使用了windows API, 去除这部分后即可用于其他平台。

mkdir build && cd build
cmake .. -G "Visual Studio 14 2015 Win64"

准备正负样本图片,其中正样本目录数据为人体图片,训练时会resize为64x128,负样本为背景图片随机抠图。 在训练时,可将负样本原图(即不包含人体)也输入,在训练过程中可进行误检困难样本的挖掘。

在trainer.cpp中设置样本路径,模型存放路径以及困难样本的保存路径,设置要训练的特征算子方法,即可开始训练。 在一轮训练后,将在负样本原图上执行算法,并将误检的图片加入训练,再进行一次训练。

测试&使用

参考detector.cpp, 输入原图,即可得到预测的检测框。如图:

展开阅读全文

About Joyk


Aggregate valuable and interesting links.
Joyk means Joy of geeK