目标检测之LibPaBOD(a LIBrary for PArt-Based Object Detection in C++)
来源:互联网 发布:sql联合主键 编辑:程序博客网 时间:2024/06/05 15:54
LibPaBOD是由Daniel Rodríguez Molina使用C++实现的DPM(P. Felzenszwalb的Discriminatively Trained Part Based Models)。代码基于opencv和MatIO实现,前者用于处理图像,后者用于读取.mat格式的训练模型文件。
1、准备工作
作者提供的windows下的可执行文件存档于\libpabod029_windows_bin\bin(LibPaBOD for MS Windows at SourceForge),包括dlls和detectobj2.exe、detectvid.exe,前者用于图片中的目标检测,后者用于视频中的目标检测。
运行程序首先需要安装hdf5:http://www.hdfgroup.org/ftp/HDF5/releases/。注意要使用shared版本,static的我尝试过没成功,所以采用的是:hdf5-1.8.13-win32-VS2008-shared.zip,然后将dll和可执行文件拷贝至相同目录:
并创建创建data目录:子目录models用于存储模型文件,子目录testimages用于存储待检测图像。
运行命令格式:
- detectobj2 -m <model_path> -i <image_path> [-t <threshold>] [-o <detections_path>] [-d <0/1>]
- detectvid -m <model_path> -v <video_path> [-t <threshold> -o <detections_path> -d <0/1>]
检测到的目标数目
目标矩形的坐标(x1 y1 x2 y2) 阈值
......
目标矩形的坐标(x1 y1 x2 y2) 阈值
2、使用作者训练的upper body detector检测头肩
头肩模型文件:ubf_libpabod.mat(upperbodyfrontal4libpabod.zip),该头肩模型来自作者的另一个VGG upper body项目。
命令如下,此处采用的阈值为-0.5:- detectobj2 -m data\models\ubf_libpabod.mat -i data\testimages\crop001007.png -t -0.5 -o ubdets.txt
结果如下:
同样检测另外几张图片:
对应的命令:
- detectobj2 -m data\models\ubf_libpabod.mat -i data\testimages\person_139.png -t -0.5 -o ubdets.txt
对应的命令:
- detectobj2 -m data\models\ubf_libpabod.mat -i data\testimages\person_269.png -t -0 -o ubdets.txt
在实验的过程中可以看出,检测依赖于阈值的选取,检测速度较慢。对比参考LatentSVM和DPM的检测:目标检测之LatentSVM和DPM(Discriminatively Trained Part Based Models) - holybin的专栏 - 博客频道 - CSDN.NET。
3、使用作者训练的人头模型检测人头
模型文件:head-gen-on-ub-4laeo.mat和head-gen-on-ub-4laeo-bmvc11.mat(laeo_head_v2.0.zip)均来自于作者的另一个项目:VGG head detector。
(1)采用第一种模型,来自:
[4] M.J. Marin-Jimenez, A. Zisserman, M. Eichner and V. Ferrari. Detecting people looking at each other in videos. International Journal of Computer Vision (IJCV), 2013
- detectobj2 -m data\models\head-gen-on-ub-4laeo.mat -i data\testimages\hlayk.jpg -t -0.82 -o ubdets.txt
检测结果:
(2)采用第二种模型,来自:
[1] M.J. Marin-Jimenez, A. Zisserman and V. Ferrari. "Here's looking at you, kid." Detecting people looking at each other in videos. British Machine Vision Conference (BMVC), 2011.
- detectobj2 -m data\models\head-gen-on-ub-4laeo-bmvc11.mat -i data\testimages\hlayk.jpg -t -0.8 -o ubdets.txt
检测结果:
4、使用人体模型检测人体
第一种模型:voc-release4.01的INRIA person模型(inriaperson_final.mat)
- detectobj2 -m data\models\inriaperson_final_4.01.mat -i data\testimages\crop001007.png -t 0.5 -o ubdets.txt
第二种模型:voc-release4.01的VOC2009 person模型(person_final.mat)
- detectobj2 -m data\models\person_final_4.01_VOC2009.mat -i data\testimages\crop001007.png -t 0.5 -o ubdets.txt
- 目标检测之LibPaBOD(a LIBrary for PArt-Based Object Detection in C++)
- 目标检测之LibPaBOD(a LIBrary for PArt-Based Object Detection in C++)
- 目标检测之LibPaBOD(a LIBrary for PArt-Based Object Detection in C++)
- LibPaBOD(a LIBrary for PArt-Based Object Detection in C++)我在运行中遇到的问题和解决方案
- 目标检测之libpabod
- (译)使用判别训练的部件模型进行目标检测 Object Detection with Discriminatively Trained Part Based Models
- Object Detection with Discriminatively Trained Part Based Models(使用判别训练的部件模型进行目标检测 )
- 目标检测“A MultiPath Network for Object Detection”
- 目标检测(Object Detection)
- 使用判别训练的部件模型进行目标检测 Object Detection with Discriminatively Trained Part Based Models
- 使用判别训练的部件模型进行目标检测 Object Detection with Discriminatively Trained Part Based Models
- 使用判别训练的部件模型进行目标检测 Object Detection with Discriminatively Trained Part Based Models
- 使用判别训练的部件模型进行目标检测 Object Detection with Discriminatively Trained Part Based Models
- 深度学习之图像目标检测(Object Detection)
- 小目标检测--A closer look: Small object detection in Faster R-CNN
- 对这个运动目标检测方法实现的结果A Hybrid Algorithm for Moving Object Detection
- 对抗学习用于目标检测--A-Fast-RCNN: Hard Positive Generation via Adversary for Object Detection
- 目标检测“A-Fast-RCNN: Hard Positive Generation via Adversary for Object Detection”
- Eclipse安装Activiti插件
- Android GPS 定位的实现(1)
- Coursera-Deep Learning Specialization 课程之(二):Improving Deep Neural Networks: -weak2编程作业
- Window10驱动简单示例(一)--文件打开读取
- Python 之旅.00
- 目标检测之LibPaBOD(a LIBrary for PArt-Based Object Detection in C++)
- javaBean简介和使用
- swiper autoheight选项卡嵌套出现空白问题
- learning R with swirl- Matrices and Data Frames(矩阵和数据帧)
- Java线程
- opencv生成点阵汉字图
- GO语言学习二(基础)
- Kubernetes高阶——设计和实现
- 投资容器云技术能赚到钱吗?