精华内容
下载资源
问答
  • OpenPose

    2020-06-13 11:11:27
    Realtime Multi-Person 2D Pose Estimation using Part Affinity Fields(Openpose) 1. https://blog.csdn.net/weixin_41441682/article/details/81357369 2. ...

    这里提出几个问题,以及整个论文流程:

    Realtime Multi-Person 2D Pose Estimation using Part Affinity Fields(Openpose)

    1. https://blog.csdn.net/weixin_41441682/article/details/81357369

    2. https://github.com/assmdx/ComputerVisionDoc/blob/master/openpose/openpose.md,中文论文网站,也就是翻译

    3. https://blog.csdn.net/htt789/article/details/80283370,这篇文章分析很透彻,能抓住问题的核心

    4. http://jacobkong.github.io/posts/3799204522/ 姿势识别论文总结

    openpose的前身就是CPM算法,相比较与CPM算法识别单人,openpose可以识别多人,虽然这么说不太严格,因为CPM算法也是可以通过center map识别出多人的,但openpose提出了PAFs,更加优雅的解决了多人的问题。

     

    è¿éåå¾çæè¿°

    这里简单介绍一下网络的流程,首先是输入一张原始图片,经过基础网络VGG进行简单的特征提取,得到一个Feature map,然后在stage1,经过两个分支分别预测。第一个分支为关键点的分支,这个分支也为CPM的经典方法,此外在这个分支的基础之上增加了PAF骨骼点走向的分支。之后的阶段与上面的类似,最终得到网络的输出S,L

    损失函数的计算,两种loss求和。

    è¿éåå¾çæè¿°       这里写图片描述

    损失函数为每一层循环网络的损失函数之和,接下来说一下两个步骤吧:

    第一个就是关键点检测,用的CPM方法。关键点检测,通过图像中标注的2D点Xj,k计算S的groundtruth(S*),其中X (j,k)表示图片中第k个人的第 j 种关节 计算方法:Sj*符合正态分布,当像素点P接近注释点Xj,k时,达到正态曲线的峰值,则每张图像中第j种关节的S,为图像中k个人的正态分布峰值。

    è¿éåå¾çæè¿°                 è¿éåå¾çæè¿°

    第二个步骤为PAF的检测:

    è¿éåå¾çæè¿°è¿éåå¾çæè¿°

    其中像素P是否落在肢体上需要满足两个条件
    这里写图片描述
    每张图像中第c中肢体的Lc*,为k个人在位置p的向量平均值
    这里写图片描述
    评估两个关键点之间的相关性:

    关键点dj1,dj2和PAF已知之后,计算两个关键点连线向量和两关键点连线上各像素的PAF向量之间的点积的积分作为两个关键点之间的相关性。
    这里写图片描述
    像素p进行采样
    这里写图片描述

    接下来就需要通过PAF计算出来的图的关系来得到一个匹配,最大匹配:匹配边数最多
    完美匹配:所有点都是匹配点
    最大权重匹配:所有匹配边权重之和最大
    最大匹配:匹配边数最多
    完美匹配:所有点都是匹配点
    最大权重匹配:所有匹配边权重之和最大
    求解最大匹配方法:匈牙利算法匈牙利算法
    https://blog.csdn.net/nineship/article/details/82797557

    展开全文
  • Openpose

    2020-12-08 18:26:02
    <div><p>Hi Sir, <p>I am very much interested in your project and want to use it for my research purpose. <p>Ubuntu: 14.04 CUDA: 8 Python: 3.6 <p>I wanted to see the demo but I have this error....
  • OPENPOSE

    2019-01-21 03:19:23
    Could NOT find CUDA (missing: CUDA_TOOLKIT_ROOT_DIR CUDA_NVCC_EXECUTABLE CUDA_INCLUDE_DIRS CUDA_CUDART_LIBRARY) https://stackoverflow.com/questions/19980412/how-to-let-cmake-find-cuda

    Could NOT find CUDA (missing: CUDA_TOOLKIT_ROOT_DIR CUDA_NVCC_EXECUTABLE CUDA_INCLUDE_DIRS CUDA_CUDART_LIBRARY)

    https://stackoverflow.com/questions/19980412/how-to-let-cmake-find-cuda

    cmake -D CUDA_TOOLKIT_ROOT_DIR=/usr/local/cuda-9.0 …
    在这里插入图片描述

    展开全文
  • openpose

    2018-11-13 10:53:00
    0 (104, 198) (70, 200) [255, 0, 0] 1 (104, 198) (138, 196) [255, 85, 0] 2 (70, 200) (60, 240) [255, 170, 0] 3 (60, 240) (42, 292) [255, 255, 0] 4 (138, 196) (168, 226) [170, 255, 0] 5 ...

    0 (104, 198) (70, 200) [255, 0, 0]
    1 (104, 198) (138, 196) [255, 85, 0]
    2 (70, 200) (60, 240) [255, 170, 0]
    3 (60, 240) (42, 292) [255, 255, 0]
    4 (138, 196) (168, 226) [170, 255, 0]
    5 (168, 226) (200, 262) [85, 255, 0]
    6 (104, 198) (94, 300) [0, 255, 0]
    7 (94, 300) (104, 360) [0, 255, 85]
    8 (104, 360) (108, 420) [0, 255, 170]
    9 (104, 198) (138, 296) [0, 255, 255]
    10 (138, 296) (148, 354) [0, 170, 255]
    11 (148, 354) (158, 406) [0, 85, 255]
    12 (104, 198) (104, 168) [0, 0, 255]
    13 (104, 168) (98, 162) [85, 0, 255]
    14 (98, 162) (86, 162) [170, 0, 255]
    15 (104, 168) (110, 160) [255, 0, 255]
    16 (110, 160) (118, 160) [255, 0, 170]

     

    转载于:https://www.cnblogs.com/herd/p/9951052.html

    展开全文
  • openpose1.7.0

    2020-12-01 11:26:26
    openpose1.7.0源码、openpose1.7.0_gpu及openpose1.7.0_cpu编译后代码(官网源码搬运分享,可自行编译VS使用)
  • OpenPose Latest News Apr 2017: Body released! May 2017: Windows version released! Jun 2017: Face released! Jul 2017: Hands released! Jul 2017: Windows, New portable demo and easier library ...
  • If you want to use OpenPose without installing or writing any code, simply download and use the latest Windows portable version of OpenPose! Otherwise, you could build OpenPose from source. See the ...
  • openpose源码

    2018-10-04 12:10:24
    openpose是姿态估计里面效果极佳的一个模型,大家都需要的话可以下载一下
  • OpenPose的安装

    千次阅读 2017-07-20 22:16:20
    OpenPose

    要求:

    Ubuntu 14/16
    GPU 2GB/1.5GB (nvidia-smi查看显存)
    2GB RAM
    8核CPU

    非常重要:
    如果安装CUDA8的ubuntu14/16:
    一.需要:CUDA, cuDNN, OpenCV, Atlas
    1.OpenCV:

    sudo apt-get install libopencv-dev

    如果已经安装OpenCV3, 修改相关makefile,把对应行改为:# OPENCV_VERSION := 3
    需要opencv_contrib模块。安装方法:

    cmake -D CMAKE_BUILD_TYPE=Release -D CMAKE_INSTALL_PREFIX=/usr/local -D OPENCV_EXTRA_MODULES_PATH=<path to opencv_contrib/modules/> ..

    例如:

    cmake -D CMAKE_BUILD_TYPE=Release -D CMAKE_INSTALL_PREFIX=/usr/local -D OPENCV_EXTRA_MODULES_PATH=/home/wsh/software/opencv-3.1.0/opencv_contrib/modules/ .. 

    然后:

    make -j7
    sudo make install

    并在makefile中的LIBRARIES += opencv_core opencv_highgui opencv_imgproc 最后加上opencv_contrib
    3.Atlas:

    sudo apt-get install libatlas-base-dev

    也可以安装OpenBLAS或Inter MKL,并将BLAS := atlas改为相应的.
    二. 安装Caffe 和 OpenPose

    chmod u+x install_caffe_and_openpose.sh
    ./install_caffe_and_openpose.sh

    如果想用CUDA7,安装Caffe:

    cd 3rdparty/caffe/
        # Select your desired Makefile file (run only one of the next 4 commands)
        cp Makefile.config.Ubuntu14_cuda_7.example Makefile.config # Ubuntu 14, cuda 7
        cp Makefile.config.Ubuntu14.example Makefile.config # Ubuntu 14, cuda 8
        cp Makefile.config.Ubuntu16_cuda_7.example Makefile.config # Ubuntu 16, cuda 7
        cp Makefile.config.Ubuntu16.example Makefile.config # Ubuntu 16, cuda 8
        # Change any custom flag from the resulting Makefile.config (e.g. OpenCV 3, Atlas/OpenBLAS/MKL, etc.)
        # Compile Caffe
        make all -j${number_of_cpus} && make distribute -j${number_of_cpus}
    ### Install OpenPose ###
        cd ../../models/
        ./getModels.sh # It just downloads the Caffe trained models
        cd ..
        # Same file cp command as the one used for Caffe
        cp Makefile.config.Ubuntu14_cuda_7.example Makefile.config
        # Change any custom flag from the resulting Makefile.config (e.g. OpenCV 3, Atlas/OpenBLAS/MKL, etc.)
        make all -j${number_of_cpus}

    如果想使用自己的Caffe,按照Custom Caffe的方法安装。
    如果你需要修改东西,请:

    make clean
    make all -j$(NUM_CORES)

    提示:
    Makefile.config.Ubuntu##.example一共有2个,其中一个在../3rdparty/caffe/中,配置文件在其中,两个文件的更改都要改:OpenCV 3 flag, Atlab/OpenBLAS/MKL flag, etc.
    重新修改的方法:

    make clean && cd 3rdparty/caffe && make clean

    然后重新生成。
    问题记录:
    1.出现:
    CXX/LD -o .build_release/tools/convert_imageset.bin
    .build_release/lib/libcaffe.so: undefined reference to cv::imread(cv::String const&, int)’ .build_release/lib/libcaffe.so: undefined reference tocv::imencode(cv::String const&, cv::_InputArray const&, std::vector >&, std::vector > const&)’
    .build_release/lib/libcaffe.so: undefined reference to `cv::imdecode(cv::_InputArray const&, int)’
    collect2: error: ld returned 1 exit status
    make: * [.build_release/tools/convert_imageset.bin] Error 1

    solution:
    add "opencv_imgcodecs" in Makefile.(LIBRARIES += glog gflags protobuf leveldb snappy \
    lmdb boost_system hdf5_hl hdf5 m \
    opencv_core opencv_highgui opencv_imgproc opencv_imgcodecs)
    If you input "make all",the problem is the same again.But if you delete all the file in build(rm -rf ./build/*) before "make all"(I use make clean ),you will success.I just success
    github 讨论帖:https://github.com/BVLC/caffe/issues/2348
    ps:
    build/lib/libcaffe.a(image_io.o): In function caffe::ReadVideoToVolumeDatum(char const*, int, int, int, int, int, int, caffe::VolumeDatum*)': image_io.cpp:(.text+0x1905): undefined reference tocv::VideoCapture::VideoCapture()'
    image_io.cpp:(.text+0x1abe): undefined reference to cv::VideoCapture::open(cv::String const&)' image_io.cpp:(.text+0x1ace): undefined reference tocv::VideoCapture::isOpened() const'
    image_io.cpp:(.text+0x1c91): undefined reference to cv::VideoCapture::~VideoCapture()' image_io.cpp:(.text+0x1d22): undefined reference tocv::VideoCapture::get(int) const'
    image_io.cpp:(.text+0x1d6d): undefined reference to cv::VideoCapture::set(int, double)' image_io.cpp:(.text+0x1de1): undefined reference tocv::VideoCapture::set(int, double)'
    image_io.cpp:(.text+0x1e35): undefined reference to cv::VideoCapture::read(cv::_OutputArray const&)' image_io.cpp:(.text+0x1f7e): undefined reference tocv::VideoCapture::release()'
    image_io.cpp:(.text+0x1fc3): undefined reference to cv::VideoCapture::read(cv::_OutputArray const&)' image_io.cpp:(.text+0x22e9): undefined reference tocv::VideoCapture::~VideoCapture()'
    collect2: error: ld returned 1 exit status
    make: *** [build/tools/blob_proto_to_blob_binary.bin] Error 1
    make: *** Waiting for unfinished jobs....
    same solution by adding opencv_videoio to LIBRARIES in the Makefile
    

    类似问题,解决方案相同。
    2.出现找不到opencv_contrib:
    换成其中有的模块
    3.出现:
    /home/usrname/opencv-3.0.0/modules/cudalegacy/src/graphcuts.cpp:120:54: error: ‘NppiGraphcutState’ has not been declared
    typedef NppStatus (init_func_t)(NppiSize oSize, NppiGraphcutState* ppStat
    /home/usrname/opencv-3.0.0/modules/cudalegacy/src/graphcuts.cpp:135:18: error: ‘NppiGraphcutState’ does not name a type
    operator NppiGraphcutState*()
    /home/usrname/opencv-3.0.0/modules/cudalegacy/src/graphcuts.cpp:141:9: error: ‘NppiGraphcutState’ does not name a type
    NppiGraphcutState* pState;
    解决方案:参考我的另外博客:
    http://blog.csdn.net/yeluohanchan/article/details/75267438
    4.出现:
    fatal error: hdf5.h: 没有那个文件或目录 :
    /opencv3.1.0/opencv_contrib/modules/hdf/include/opencv2/hdf/hdf5.hpp
    把:

    #include <hdf5.h>

    改为:

    #include </usr/include/hdf5/serial/hdf5.h>

    5.运行出错:
    ./build/examples/openpose/rtpose.bin: error while loading shared libraries: libopencv_core.so.3.1: cannot open shared object file: No such file or directory
    解决方法:

    sudo ldconfig
    
    展开全文
  • OpenPose Demo

    2020-09-04 10:41:29
    OpenPose Demo 忘记OpenPose库代码,只编译库并使用演示./build/examples/openpose/openpose.bin
  • OpenPose安装

    2020-09-03 11:14:47
    目录Windows Portable Demo安装视频教程操作系统Community-Based Work依赖软件克隆OpenPose更新OpenPose安装安装过程中出现的问题Prerequisites配置OpenPose构建OpenPose运行OpenPoseWindows说明重装卸载部署...
  • openpose_win

    2020-10-31 21:10:03
    openpose在Windows下编译需要的caffe和opencv的zip包,cmake编译时下载很慢,放在openpose-master\3rdparty\windows下再进行cmake编译
  • OpenPose笔记PPT

    2018-08-15 09:38:49
    openpose动作检测笔记,详细解释openpose检测的各个细节,图文并茂。
  • Lightweight OpenPose

    2020-09-28 15:58:05
    Real-time 2D Multi-Person Pose Estimation on CPU: Lightweight OpenPose 1、优势 1) 迷你PCIntel®NUC上的28 fps(它消耗的功率很少,并具有45瓦CPU TDP) 2)在常规CPU上为26 fps,而无需图形处理器 3)AP精度...
  • openpose源文章

    2018-07-31 20:38:07
    包含openpose原理的三篇原始文章,帮助大家了解openpose的来历,是值得收藏的好文章,欢迎大家下载
  • openpose-models

    2020-08-27 17:47:36
    在windows10下,openpose中的models文件,models文件有:pose_iter_102000.caffemodel;pose_iter_116000.caffemodel;pose_iter_160000.caffemodel;pose_iter_440000.caffemodel;pose_iter_584000.caffemodel。
  • caffe-openpose

    2019-04-11 18:30:09
    鉴于在github上下载速度过慢的原因,因此将该code上传到csdn上,可以用于openpose的训练
  • Openpose实操

    2020-06-29 15:44:51
    一,OpenPose项目Github链接:https://github.com/CMU-Perceptual-Computing-Lab/openpose 二,windows源码下载地址: 源码下载 下载第一个推荐版。 三,下载训练好的caffe模型文件 百度云盘上供大家下载使用: ...
  • openpose_tennis-源码

    2021-02-18 09:02:37
    openpose_tennis
  • OpenPose Unity插件是Unity用户的库的包装。 它提供了格式化的OpenPose输出和一些示例。 是一个实时多人系统,能够共同检测单个图像上的人体,手,面部和脚的关键点(总共135个关键点)。 OpenPose Unity插件项目由...
  • openpose模型:hand

    2019-01-30 17:34:06
    openpose模型:hand
  • openpose最新下载

    2018-06-14 16:43:54
    OpenPose is a library for real-time multi-person keypoint detection and multi-threading written in C++ using OpenCV and Caffe.
  • 一套处理流程,适用于多种数据类型,实现代码复用.openpose这个项目所包含的不仅仅是姿态估计/人脸/手势关键点的定位算法;它所实现的一套多线程框架代码(用于组织各种任务的协同运行)同样也非常值得学习,由于这...
  • opencv-openpose.zip

    2021-02-23 18:01:18
    opencv-openpose.zip
  • darknet+openpose

    2018-06-04 11:50:15
    修改过的darknet,版本为yolo2,如果有需要c++,cmake版yolo3可联系yeluohanchan@126.com。这个是darknet+openpose的c++版本,用cmake编译,操作系统是ubuntu16.04, opencv3.1+cuda8.0+cudnn5.1

空空如也

空空如也

1 2 3 4 5 ... 20
收藏数 1,941
精华内容 776
关键字:

openpose