Hi,
i see the openpose install instruction only for Jetpack 3.3 https://github.com/CMU-Perceptual-Computing-Lab/openpose/blob/master/doc/installation_jetson_tx2_jetpack3.3.md
Is there any guide/help for Jetpack 4.2 on Jetson TX2?
Please help.
Hi,
Here are the steps for your reference:
1. Install cmake
$ wget http://www.cmake.org/files/v3.12/cmake-3.12.2.tar.gz
$ tar -xvzf cmake-3.12.2.tar.gz
$ cd cmake-3.12.2/
$ ./configure
$ make
2. Install dependencies
$ sudo apt-get install libboost-dev libboost-all-dev
$ sudo apt-get install libgflags-dev libgoogle-glog-dev liblmdb-dev libatlas-base-dev liblmdb-dev libblas-dev libatlas-base-dev libprotobuf-dev libleveldb-dev libsnappy-dev libhdf5-serial-dev protobuf-compiler
3. Install openpose
$ git clone --recursive https://github.com/CMU-Perceptual-Computing-Lab/openpose
$ mkdir build
[Xavier] $ /home/nvidia/cmake-3.12.2/bin/cmake ../ -DCUDA_ARCH_BIN=53
[TX2] $ /home/nvidia/cmake-3.12.2/bin/cmake ../ -DCUDA_ARCH_BIN=62
[TX1] $ /home/nvidia/cmake-3.12.2/bin/cmake ../ -DCUDA_ARCH_BIN=53
[Nano] $ /home/nvidia/cmake-3.12.2/bin/cmake ../ -DCUDA_ARCH_BIN=53
$ make -j4
$ sudo make install
4. Run
./build/examples/openpose/openpose.bin -camera_resolution 640x480 -net_resolution 128x96
Thanks.
Hi,
thanks for your help. I was able to setup and run openpose with Realsense D435i connected and here is the output. With this i can see up to 15FPS which is really high,
jetson@jetson-desktop: ~/openpose
$ ./build/examples/openpose/openpose.bin -camera_resolution 640x480 -net_resolution 128x96
Starting OpenPose demo…
Configuring OpenPose…
Starting thread(s)…
VIDEOIO ERROR: V4L2: Pixel format of incoming image is unsupported by OpenCV
OpenCV Error: Unspecified error (GStreamer: unable to start pipeline
) in cvCaptureFromCAM_GStreamer, file /home/nvidia/build_opencv/opencv/modules/videoio/src/cap_gstreamer.cpp, line 887
VIDEOIO(cvCreateCapture_GStreamer(CV_CAP_GSTREAMER_V4L2, reinterpret_cast<char *>(index))): raised OpenCV exception:
/home/nvidia/build_opencv/opencv/modules/videoio/src/cap_gstreamer.cpp:887: error: (-2) GStreamer: unable to start pipeline
in function cvCaptureFromCAM_GStreamer
VIDEOIO ERROR: V4L2: Pixel format of incoming image is unsupported by OpenCV
OpenCV Error: Unspecified error (GStreamer: unable to start pipeline
) in cvCaptureFromCAM_GStreamer, file /home/nvidia/build_opencv/opencv/modules/videoio/src/cap_gstreamer.cpp, line 887
VIDEOIO(cvCreateCapture_GStreamer(CV_CAP_GSTREAMER_V4L2, reinterpret_cast<char *>(index))): raised OpenCV exception:
/home/nvidia/build_opencv/opencv/modules/videoio/src/cap_gstreamer.cpp:887: error: (-2) GStreamer: unable to start pipeline
in function cvCaptureFromCAM_GStreamer
Auto-detecting camera index… Detected and opened camera 2.
Auto-detecting all available GPUs… Detected 1 GPU(s), using 1 of them starting at GPU 0.
Gtk-Message: 09:47:10.650: Failed to load module “canberra-gtk-module”
OpenPose demo successfully finished. Total time: 81.616844 seconds.
jetson@jetson-desktop: ~/openpose
But how do we fix these errors from VIDEOIO and Opencv?
Again i am not sure why is it referring to /home/nvidia? whereas my home directory name is /home/jetson.
is there any rule that home directory name should be “nvidia”?
Hi,
The error is from OpenCV+GStreamer.
Please noticed that our default OpenCV package doesn’t enable GStreamer support.
You will need to build it from source to launch a GStreamer pipeline within OpenCV.
Here is our building script for your reference:
https://github.com/AastaNV/JEP/blob/master/script/install_opencv4.0.0_Nano.sh
Update this command for TX2 GPU capacity:
cmake -D WITH_CUDA=ON -D CUDA_ARCH_BIN="6.2" -D CUDA_ARCH_PTX="" -D OPENCV_EXTRA_MODULES_PATH=../../opencv_contrib-4.0.0/modules -D WITH_GSTREAMER=ON -D WITH_LIBV4L=ON -D BUILD_opencv_python2=ON -D BUILD_opencv_python3=ON -D BUILD_TESTS=OFF -D BUILD_PERF_TESTS=OFF -D BUILD_EXAMPLES=OFF -D CMAKE_BUILD_TYPE=RELEASE -D CMAKE_INSTALL_PREFIX=/usr/local ..
Thanks.
Hi,
After i install opencv 4.0 as you said, i had to recompile openpose and run again. But this time it is not even launch video. can you help what is the issue heree?
jetson@jetson-desktop: ~/openpose
$ ./build/examples/openpose/openpose.bin -camera_resolution 640x480 -net_resolution 128x96
Starting OpenPose demo…
Configuring OpenPose…
Starting thread(s)…
(openpose.bin:26241): GStreamer-CRITICAL **: 13:57:07.450: gst_element_get_state: assertion ‘GST_IS_ELEMENT (element)’ failed
VIDEOIO ERROR: V4L2: Pixel format of incoming image is unsupported by OpenCV
VIDEOIO ERROR: V4L: can’t open camera by index 0
(openpose.bin:26241): GStreamer-CRITICAL **: 13:57:07.472: gst_element_get_state: assertion ‘GST_IS_ELEMENT (element)’ failed
VIDEOIO ERROR: V4L2: Pixel format of incoming image is unsupported by OpenCV
VIDEOIO ERROR: V4L: can’t open camera by index 1
Desired webcam resolution 640x480 could not being set. Final resolution: 1280x720 in /home/jetson/openpose/src/openpose/producer/webcamReader.cpp:WebcamReader():36
Auto-detecting camera index… Detected and opened camera 2.
Auto-detecting all available GPUs… Detected 1 GPU(s), using 1 of them starting at GPU 0.
Error:
OpenCV(4.0.0) /home/jetson/opencv-4.0.0/modules/highgui/src/window_gtk.cpp:146: error: (-215:Assertion failed) dst.data == widget->original_image->data.ptr in function ‘cvImageWidgetSetImage’
Coming from:
- /home/jetson/openpose/src/openpose/gui/frameDisplayer.cpp:displayFrame():116
- /home/jetson/openpose/src/openpose/gui/frameDisplayer.cpp:initializationOnThread():44
- /home/jetson/openpose/include/openpose/gui/wGui.hpp:initializationOnThread():57
- /home/jetson/openpose/include/openpose/thread/worker.hpp:initializationOnThreadNoException():77
- /home/jetson/openpose/include/openpose/thread/subThread.hpp:initializationOnThread():150
- /home/jetson/openpose/include/openpose/thread/thread.hpp:initializationOnThread():173
- /home/jetson/openpose/include/openpose/thread/thread.hpp:threadFunction():203
- /home/jetson/openpose/include/openpose/thread/thread.hpp:exec():128
- /home/jetson/openpose/include/openpose/thread/threadManager.hpp:exec():202
- /home/jetson/openpose/include/openpose/wrapper/wrapper.hpp:exec():424
jetson@jetson-desktop: ~/openpose
Hi,
May I know your camera type? Is it CSI camera or USB camera?
Could you also help to check if it can be opened with GStreamer or not first?
gst-launch-1.0 v4l2src device="/dev/video0" ! "video/x-raw, width=640, height=480, format=(string)YUY2" ! xvimagesink -e
Thanks.
Hi,
Followed the steps above, but my jetson tx2 automatically restarts after trying to get to my logitech webcam. I’ve gotten openpose to work on jetpack 3.3 so I know it works, but for 4.2 something else seems to be happening. I’d like some clues as to why my entire jetson tx2 restarts after i run a video command from the explanation above.
I’ve followed every single line to the T and everything seemed to install correctly.
Hi AastaLLL,
It works fine for me after fixing the camera id. thanks.
It looks like it installed correctly and the cmake also looks like it built and installed caffe but I get the following errors when trying to run the 01_body_from_image_default.bin example binary.
Error:
Prototxt file not found: models/pose/body_25/pose_deploy.prototxt.
Possible causes:
1. Not downloading the OpenPose trained models.
2. Not running OpenPose from the same directory where the `model` folder is located.
3. Using paths with spaces.
Coming from:
- /home/nvidia/datacapture/openpose/src/openpose/net/netCaffe.cpp:ImplNetCaffe():58
- /home/nvidia/datacapture/openpose/src/openpose/net/netCaffe.cpp:ImplNetCaffe():94
- /home/nvidia/datacapture/openpose/src/openpose/pose/poseExtractorCaffe.cpp:addCaffeNetOnThread():106
- /home/nvidia/datacapture/openpose/src/openpose/pose/poseExtractorCaffe.cpp:netInitializationOnThread():196
- /home/nvidia/datacapture/openpose/src/openpose/pose/poseExtractorNet.cpp:initializationOnThread():99
- /home/nvidia/datacapture/openpose/src/openpose/pose/poseExtractor.cpp:initializationOnThread():34
- /home/nvidia/datacapture/openpose/include/openpose/pose/wPoseExtractor.hpp:initializationOnThread():57
- /home/nvidia/datacapture/openpose/include/openpose/thread/worker.hpp:initializationOnThreadNoException():77
Error:
Wrong input element (empty cvInputData).
Coming from:
- /home/nvidia/datacapture/openpose/src/openpose/core/scaleAndSizeExtractor.cpp:extract():43
- /home/nvidia/datacapture/openpose/src/openpose/core/scaleAndSizeExtractor.cpp:extract():99
- /home/nvidia/datacapture/openpose/include/openpose/core/wScaleAndSizeExtractor.hpp:work():83
- /home/nvidia/datacapture/openpose/include/openpose/thread/worker.hpp:checkAndWork():93
Body keypoints: Array<T>::toString():
Error:
OpenCV(3.4.6) /home/nvidia/datacapture/src/opencv-3.4.6/modules/highgui/src/window.cpp:366: error: (-215:Assertion failed) size.width>0 && size.height>0 in function 'imshow'
Coming from:
- /home/nvidia/datacapture/openpose/examples/tutorial_api_cpp/01_body_from_image_default.cpp:display():40
Error occurred on a thread. OpenPose closed all its threads and then propagated the error to the main thread. Error description:
Prototxt file not found: models/pose/body_25/pose_deploy.prototxt.
Possible causes:
1. Not downloading the OpenPose trained models.
2. Not running OpenPose from the same directory where the `model` folder is located.
3. Using paths with spaces.
I just tried running it from the openpose base folder using the following command:
nvidia@nvidia-desktop:~/datacapture/openpose$ ./build/examples/tutorial_api_cpp/01_body_from_image_default.bin
It will run the first few comments until it gets to the GPU inquiry and afterwards when it appears as though its trying to run the code it immediately kills my TX2 and powers off? Could this be something with the Caffe buid? I rebuilt my OpenCV to 3.4.6 with GStreamer capabilities ON since the 3.3.1 version does not come with GST ON. I have tried using the GUI as well as manual cmake installation through the terminal as I did this and got it working on JetPack 3.3 awhile back.
Here are my CMakeList.txt Configure and Generate Results as well:
GCC detected, adding compile flags
GCC detected, adding compile flags
Building with CUDA.
eigen_2018_05_23.zip already exists.
CUDA detected: 10.0
Found cuDNN: ver. 7.3.1 found (include: /usr/include, library: /usr/lib/aarch64-linux-gnu/libcudnn.so)
Added CUDA NVCC flags for: sm_62
Found cuDNN: ver. 7.3.1 found (include: /usr/include, library: /usr/lib/aarch64-linux-gnu/libcudnn.so)
Found gflags (include: /usr/include, library: /usr/lib/aarch64-linux-gnu/libgflags.so)
Found glog (include: /usr/include, library: /usr/lib/aarch64-linux-gnu/libglog.so)
Caffe will be downloaded from source now. NOTE: This process might take several minutes depending
on your internet connection.
Caffe has already been downloaded.
M .Doxyfile
M .github/ISSUE_TEMPLATE.md
M .gitignore
M .travis.yml
M CMakeLists.txt
M CONTRIBUTING.md
M CONTRIBUTORS.md
M INSTALL.md
M LICENSE
M Makefile
M Makefile.config.Ubuntu14_cuda7.example
M Makefile.config.Ubuntu14_cuda8.example
M Makefile.config.Ubuntu16_cuda7.example
M Makefile.config.Ubuntu16_cuda8.example
M Makefile.config.Ubuntu16_cuda8_JetsonTX2
M Makefile.config.Ubuntu16_cuda9_JetsonTX2_JetPack33
M README.md
M caffe.cloc
M cmake/ConfigGen.cmake
M cmake/Cuda.cmake
M cmake/Dependencies.cmake
M cmake/External/gflags.cmake
M cmake/External/glog.cmake
M cmake/Misc.cmake
M cmake/Modules/FindAtlas.cmake
M cmake/Modules/FindGFlags.cmake
M cmake/Modules/FindGlog.cmake
M cmake/Modules/FindLAPACK.cmake
M cmake/Modules/FindLMDB.cmake
M cmake/Modules/FindLevelDB.cmake
M cmake/Modules/FindMKL.cmake
M cmake/Modules/FindMatlabMex.cmake
M cmake/Modules/FindNCCL.cmake
M cmake/Modules/FindNumPy.cmake
M cmake/Modules/FindOpenBLAS.cmake
M cmake/Modules/FindSnappy.cmake
M cmake/Modules/FindvecLib.cmake
M cmake/ProtoB
uf.cmake
M cmake/Summary.cmake
M cmake/Targets.cmake
M cmake/Templates/CaffeConfig.cmake.in
M cmake/Templates/CaffeConfigVersion.cmake.in
M cmake/Templates/caffe_config.h.in
M cmake/Uninstall.cmake.in
M cmake/Utils.cmake
M cmake/lint.cmake
M docker/README.md
M docker/cpu/Dockerfile
M docker/gpu/Dockerfile
M docs/CMakeLists.txt
M docs/CNAME
M docs/README.md
M docs/_config.yml
M docs/_layouts/default.html
M docs/development.md
M docs/images/GitHub-Mark-64px.png
M docs/images/caffeine-icon.png
M docs/index.md
M docs/install_apt.md
M docs/install_apt_debian.md
M docs/install_osx.md
M docs/install_yum.md
M docs/installation.md
M docs/model_zoo.md
M docs/multigpu.md
M docs/stylesheets/pygment_trac.css
M docs/stylesheets/reset.css
M docs/stylesheets/styles.css
M docs/tutorial/convolution.md
M docs/tutorial/data.md
M docs/tutorial/fig/.gitignore
M docs/tutorial/fig/backward.jpg
M docs/tutorial/fig/forward.jpg
M docs/tutorial/fig/forward_backward.png
M docs/tutorial/fig/layer.jpg
M docs/tutorial/fig/logreg.jpg
M docs/
tutorial/forward_backward.md
M docs/tutorial/index.md
M docs/tutorial/interfaces.md
M docs/tutorial/layers.md
M docs/tutorial/layers/absval.md
M docs/tutorial/layers/accuracy.md
M docs/tutorial/layers/argmax.md
M docs/tutorial/layers/batchnorm.md
M docs/tutorial/layers/batchreindex.md
M docs/tutorial/layers/bias.md
M docs/tutorial/layers/bnll.md
M docs/tutorial/layers/clip.md
M docs/tutorial/layers/concat.md
M docs/tutorial/layers/contrastiveloss.md
M docs/tutorial/layers/convolution.md
M docs/tutorial/layers/crop.md
M docs/tutorial/layers/data.md
M docs/tutorial/layers/deconvolution.md
M docs/tutorial/layers/dropout.md
M docs/tutorial/layers/dummydata.md
M docs/tutorial/layers/eltwise.md
M docs/tutorial/layers/elu.md
M docs/tutorial/layers/embed.md
M docs/tutorial/layers/euclideanloss.md
M docs/tutorial/layers/exp.md
M docs/tutorial/layers/filter.md
M docs/tutorial/layers/flatten.md
M docs/tutorial/layers/hdf5data.md
M docs/tutorial/layers/hdf5output.md
M docs/tutorial/layers/hingeloss.md
M docs/tutorial/lay
ers/im2col.md
M docs/tutorial/layers/imagedata.md
M docs/tutorial/layers/infogainloss.md
M docs/tutorial/layers/innerproduct.md
M docs/tutorial/layers/input.md
M docs/tutorial/layers/log.md
M docs/tutorial/layers/lrn.md
M docs/tutorial/layers/lstm.md
M docs/tutorial/layers/memorydata.md
M docs/tutorial/layers/multinomiallogisticloss.md
M docs/tutorial/layers/mvn.md
M docs/tutorial/layers/parameter.md
M docs/tutorial/layers/pooling.md
M docs/tutorial/layers/power.md
M docs/tutorial/layers/prelu.md
M docs/tutorial/layers/python.md
M docs/tutorial/layers/recurrent.md
M docs/tutorial/layers/reduction.md
M docs/tutorial/layers/relu.md
M docs/tutorial/layers/reshape.md
M docs/tutorial/layers/rnn.md
M docs/tutorial/layers/scale.md
M docs/tutorial/layers/sigmoid.md
M docs/tutorial/layers/sigmoidcrossentropyloss.md
M docs/tutorial/layers/silence.md
M docs/tutorial/layers/slice.md
M docs/tutorial/layers/softmax.md
M docs/tutorial/layers/softmaxwithloss.md
M docs/tutorial/layers/split.md
M docs/tutorial/layers/spp.md
M
docs/tutorial/layers/tanh.md
M docs/tutorial/layers/threshold.md
M docs/tutorial/layers/tile.md
M docs/tutorial/layers/windowdata.md
M docs/tutorial/loss.md
M docs/tutorial/net_layer_blob.md
M docs/tutorial/solver.md
M examples/00-classification.ipynb
M examples/01-learning-lenet.ipynb
M examples/02-fine-tuning.ipynb
M examples/CMakeLists.txt
M examples/brewing-logreg.ipynb
M examples/cifar10/cifar10_full.prototxt
M examples/cifar10/cifar10_full_sigmoid_solver.prototxt
M examples/cifar10/cifar10_full_sigmoid_solver_bn.prototxt
M examples/cifar10/cifar10_full_sigmoid_train_test.prototxt
M examples/cifar10/cifar10_full_sigmoid_train_test_bn.prototxt
M examples/cifar10/cifar10_full_solver.prototxt
M examples/cifar10/cifar10_full_solver_lr1.prototxt
M examples/cifar10/cifar10_full_solver_lr2.prototxt
M examples/cifar10/cifar10_full_train_test.prototxt
M examples/cifar10/cifar10_quick.prototxt
M examples/cifar10/cifar10_quick_solver.prototxt
M examples/cifar10/cifar10_quick_solver_lr1.prototxt
M examples/cifar10/c
ifar10_quick_train_test.prototxt
M examples/cifar10/convert_cifar_data.cpp
M examples/cifar10/readme.md
M examples/cpp_classification/classification.cpp
M examples/cpp_classification/readme.md
M examples/detection.ipynb
M examples/feature_extraction/imagenet_val.prototxt
M examples/feature_extraction/readme.md
M examples/finetune_flickr_style/flickr_style.csv.gz
M examples/finetune_flickr_style/readme.md
M examples/finetune_flickr_style/style_names.txt
M examples/finetune_pascal_detection/pascal_finetune_solver.prototxt
M examples/finetune_pascal_detection/pascal_finetune_trainval_test.prototxt
M examples/hdf5_classification/nonlinear_auto_test.prototxt
M examples/hdf5_classification/nonlinear_auto_train.prototxt
M examples/hdf5_classification/nonlinear_train_val.prototxt
M examples/hdf5_classification/train_val.prototxt
M examples/imagenet/readme.md
M examples/images/cat gray.jpg
M examples/images/cat.jpg
M examples/images/cat_gray.jpg
M examples/images/fish-bike.jpg
M examples/mnist/convert_mnist_data.cpp
M
examples/mnist/lenet.prototxt
M examples/mnist/lenet_adadelta_solver.prototxt
M examples/mnist/lenet_auto_solver.prototxt
M examples/mnist/lenet_consolidated_solver.prototxt
M examples/mnist/lenet_multistep_solver.prototxt
M examples/mnist/lenet_solver.prototxt
M examples/mnist/lenet_solver_adam.prototxt
M examples/mnist/lenet_solver_rmsprop.prototxt
M examples/mnist/lenet_train_test.prototxt
M examples/mnist/mnist_autoencoder.prototxt
M examples/mnist/mnist_autoencoder_solver.prototxt
M examples/mnist/mnist_autoencoder_solver_adadelta.prototxt
M examples/mnist/mnist_autoencoder_solver_adagrad.prototxt
M examples/mnist/mnist_autoencoder_solver_nesterov.prototxt
M examples/mnist/readme.md
M examples/net_surgery.ipynb
M examples/net_surgery/bvlc_caffenet_full_conv.prototxt
M examples/net_surgery/conv.prototxt
M examples/pascal-multilabel-with-datalayer.ipynb
M examples/pycaffe/caffenet.py
M examples/pycaffe/layers/pascal_multilabel_datalayers.py
M examples/pycaffe/layers/pyloss.py
M examples/pycaffe/linreg.pro
totxt
M examples/pycaffe/tools.py
M examples/siamese/convert_mnist_siamese_data.cpp
M examples/siamese/mnist_siamese.ipynb
M examples/siamese/mnist_siamese.prototxt
M examples/siamese/mnist_siamese_solver.prototxt
M examples/siamese/mnist_siamese_train_test.prototxt
M examples/siamese/readme.md
M examples/web_demo/app.py
M examples/web_demo/exifutil.py
M examples/web_demo/readme.md
M examples/web_demo/requirements.txt
M examples/web_demo/templates/index.html
M include/caffe/blob.hpp
M include/caffe/caffe.hpp
M include/caffe/common.hpp
M include/caffe/data_transformer.hpp
M include/caffe/filler.hpp
M include/caffe/internal_thread.hpp
M include/caffe/layer.hpp
M include/caffe/layer_factory.hpp
M include/caffe/layers/absval_layer.hpp
M include/caffe/layers/accuracy_layer.hpp
M include/caffe/layers/argmax_layer.hpp
M include/caffe/layers/base_conv_layer.hpp
M include/caffe/layers/base_data_layer.hpp
M include/caffe/layers/batch_norm_layer.hpp
M include/caffe/layers/batch_reindex_layer.hpp
M include/caffe/layers/b
ias_layer.hpp
M include/caffe/layers/bnll_layer.hpp
M include/caffe/layers/clip_layer.hpp
M include/caffe/layers/concat_layer.hpp
M include/caffe/layers/contrastive_loss_layer.hpp
M include/caffe/layers/conv_layer.hpp
M include/caffe/layers/crop_layer.hpp
M include/caffe/layers/cudnn_conv_layer.hpp
M include/caffe/layers/cudnn_deconv_layer.hpp
M include/caffe/layers/cudnn_lcn_layer.hpp
M include/caffe/layers/cudnn_lrn_layer.hpp
M include/caffe/layers/cudnn_pooling_layer.hpp
M include/caffe/layers/cudnn_relu_layer.hpp
M include/caffe/layers/cudnn_sigmoid_layer.hpp
M include/caffe/layers/cudnn_softmax_layer.hpp
M include/caffe/layers/cudnn_tanh_layer.hpp
M include/caffe/layers/data_layer.hpp
M include/caffe/layers/deconv_layer.hpp
M include/caffe/layers/dropout_layer.hpp
M include/caffe/layers/dummy_data_layer.hpp
M include/caffe/layers/eltwise_layer.hpp
M include/caffe/layers/elu_layer.hpp
M include/caffe/layers/embed_layer.hpp
M include/caffe/layers/euclidean_loss_layer.hpp
M include/caffe/layers/exp_layer.hp
p
M include/caffe/layers/filter_layer.hpp
M include/caffe/layers/flatten_layer.hpp
M include/caffe/layers/hdf5_data_layer.hpp
M include/caffe/layers/hdf5_output_layer.hpp
M include/caffe/layers/hinge_loss_layer.hpp
M include/caffe/layers/im2col_layer.hpp
M include/caffe/layers/image_data_layer.hpp
M include/caffe/layers/infogain_loss_layer.hpp
M include/caffe/layers/inner_product_layer.hpp
M include/caffe/layers/input_layer.hpp
M include/caffe/layers/log_layer.hpp
M include/caffe/layers/loss_layer.hpp
M include/caffe/layers/lrn_layer.hpp
M include/caffe/layers/lstm_layer.hpp
M include/caffe/layers/memory_data_layer.hpp
M include/caffe/layers/multinomial_logistic_loss_layer.hpp
M include/caffe/layers/mvn_layer.hpp
M include/caffe/layers/neuron_layer.hpp
M include/caffe/layers/parameter_layer.hpp
M include/caffe/layers/pooling_layer.hpp
M include/caffe/layers/power_layer.hpp
M include/caffe/layers/prelu_layer.hpp
M include/caffe/layers/python_layer.hpp
M include/caffe/layers/recurrent_layer.hpp
M include/caffe/
layers/reduction_layer.hpp
M include/caffe/layers/relu_layer.hpp
M include/caffe/layers/reshape_layer.hpp
M include/caffe/layers/rnn_layer.hpp
M include/caffe/layers/scale_layer.hpp
M include/caffe/layers/sigmoid_cross_entropy_loss_layer.hpp
M include/caffe/layers/sigmoid_layer.hpp
M include/caffe/layers/silence_layer.hpp
M include/caffe/layers/slice_layer.hpp
M include/caffe/layers/softmax_layer.hpp
M include/caffe/layers/softmax_loss_layer.hpp
M include/caffe/layers/split_layer.hpp
M include/caffe/layers/spp_layer.hpp
M include/caffe/layers/swish_layer.hpp
M include/caffe/layers/tanh_layer.hpp
M include/caffe/layers/threshold_layer.hpp
M include/caffe/layers/tile_layer.hpp
M include/caffe/layers/window_data_layer.hpp
M include/caffe/net.hpp
M include/caffe/parallel.hpp
M include/caffe/sgd_solvers.hpp
M include/caffe/solver.hpp
M include/caffe/solver_factory.hpp
M include/caffe/syncedmem.hpp
M include/caffe/test/test_caffe_main.hpp
M include/caffe/test/test_gradient_check_util.hpp
M include/caffe/util/benchm
ark.hpp
M include/caffe/util/blocking_queue.hpp
M include/caffe/util/cudnn.hpp
M include/caffe/util/db.hpp
M include/caffe/util/db_leveldb.hpp
M include/caffe/util/db_lmdb.hpp
M include/caffe/util/device_alternate.hpp
M include/caffe/util/format.hpp
M include/caffe/util/gpu_util.cuh
M include/caffe/util/hdf5.hpp
M include/caffe/util/im2col.hpp
M include/caffe/util/insert_splits.hpp
M include/caffe/util/io.hpp
M include/caffe/util/math_functions.hpp
M include/caffe/util/mkl_alternate.hpp
M include/caffe/util/nccl.hpp
M include/caffe/util/rng.hpp
M include/caffe/util/signal_handler.h
M include/caffe/util/upgrade_proto.hpp
M install_caffe_JetsonTX2_JetPack3.3.sh
M matlab/+caffe/+test/test_io.m
M matlab/+caffe/+test/test_net.m
M matlab/+caffe/+test/test_solver.m
M matlab/+caffe/Blob.m
M matlab/+caffe/Layer.m
M matlab/+caffe/Net.m
M matlab/+caffe/Solver.m
M matlab/+caffe/get_net.m
M matlab/+caffe/get_solver.m
M matlab/+caffe/imagenet/ilsvrc_2012_mean.mat
M matlab/+caffe/io.m
M matlab/+caffe/private/CHECK.m
M matla
b/+caffe/private/CHECK_FILE_EXIST.m
M matlab/+caffe/private/caffe_.cpp
M matlab/+caffe/private/is_valid_handle.m
M matlab/+caffe/reset_all.m
M matlab/+caffe/run_tests.m
M matlab/+caffe/set_device.m
M matlab/+caffe/set_mode_cpu.m
M matlab/+caffe/set_mode_gpu.m
M matlab/+caffe/version.m
M matlab/CMakeLists.txt
M matlab/demo/classification_demo.m
M matlab/hdf5creation/.gitignore
M matlab/hdf5creation/demo.m
M matlab/hdf5creation/store2hdf5.m
M models/bvlc_alexnet/deploy.prototxt
M models/bvlc_alexnet/readme.md
M models/bvlc_alexnet/solver.prototxt
M models/bvlc_alexnet/train_val.prototxt
M models/bvlc_googlenet/deploy.prototxt
M models/bvlc_googlenet/quick_solver.prototxt
M models/bvlc_googlenet/readme.md
M models/bvlc_googlenet/solver.prototxt
M models/bvlc_reference_caffenet/deploy.prototxt
M models/bvlc_reference_caffenet/readme.md
M models/bvlc_reference_caffenet/solver.prototxt
M models/bvlc_reference_caffenet/train_val.prototxt
M models/bvlc_reference_rcnn_ilsvrc13/deploy.prototxt
M models/bvlc_reference_r
cnn_ilsvrc13/readme.md
M models/finetune_flickr_style/deploy.prototxt
M models/finetune_flickr_style/readme.md
M models/finetune_flickr_style/solver.prototxt
M models/finetune_flickr_style/train_val.prototxt
M python/CMakeLists.txt
M python/caffe/__init__.py
M python/caffe/_caffe.cpp
M python/caffe/classifier.py
M python/caffe/coord_map.py
M python/caffe/detector.py
M python/caffe/draw.py
M python/caffe/imagenet/ilsvrc_2012_mean.npy
M python/caffe/io.py
M python/caffe/net_spec.py
M python/caffe/pycaffe.py
M python/caffe/test/test_coord_map.py
M python/caffe/test/test_draw.py
M python/caffe/test/test_io.py
M python/caffe/test/test_layer_type_list.py
M python/caffe/test/test_nccl.py
M python/caffe/test/test_net.py
M python/caffe/test/test_net_spec.py
M python/caffe/test/test_python_layer.py
M python/caffe/test/test_python_layer_with_param_str.py
M python/caffe/test/test_solver.py
M python/requirements.txt
M python/train.py
M scripts/caffe
M scripts/travis/configure-cmake.sh
M scripts/travis/configure-make.sh
M
src/caffe/CMakeLists.txt
M src/caffe/blob.cpp
M src/caffe/common.cpp
M src/caffe/data_transformer.cpp
M src/caffe/internal_thread.cpp
M src/caffe/layer.cpp
M src/caffe/layer_factory.cpp
M src/caffe/layers/absval_layer.cpp
M src/caffe/layers/absval_layer.cu
M src/caffe/layers/accuracy_layer.cpp
M src/caffe/layers/accuracy_layer.cu
M src/caffe/layers/argmax_layer.cpp
M src/caffe/layers/base_conv_layer.cpp
M src/caffe/layers/base_data_layer.cpp
M src/caffe/layers/base_data_layer.cu
M src/caffe/layers/batch_norm_layer.cpp
M src/caffe/layers/batch_norm_layer.cu
M src/caffe/layers/batch_reindex_layer.cpp
M src/caffe/layers/batch_reindex_layer.cu
M src/caffe/layers/bias_layer.cpp
M src/caffe/layers/bias_layer.cu
M src/caffe/layers/bnll_layer.cpp
M src/caffe/layers/bnll_layer.cu
M src/caffe/layers/clip_layer.cpp
M src/caffe/layers/clip_layer.cu
M src/caffe/layers/concat_layer.cpp
M src/caffe/layers/concat_layer.cu
M src/caffe/layers/contrastive_loss_layer.cpp
M src/caffe/layers/contrastive_loss_layer.cu
M src/caffe/l
ayers/conv_layer.cpp
M src/caffe/layers/conv_layer.cu
M src/caffe/layers/crop_layer.cpp
M src/caffe/layers/crop_layer.cu
M src/caffe/layers/cudnn_conv_layer.cpp
M src/caffe/layers/cudnn_conv_layer.cu
M src/caffe/layers/cudnn_deconv_layer.cpp
M src/caffe/layers/cudnn_deconv_layer.cu
M src/caffe/layers/cudnn_lcn_layer.cpp
M src/caffe/layers/cudnn_lcn_layer.cu
M src/caffe/layers/cudnn_lrn_layer.cpp
M src/caffe/layers/cudnn_lrn_layer.cu
M src/caffe/layers/cudnn_pooling_layer.cpp
M src/caffe/layers/cudnn_pooling_layer.cu
M src/caffe/layers/cudnn_relu_layer.cpp
M src/caffe/layers/cudnn_relu_layer.cu
M src/caffe/layers/cudnn_sigmoid_layer.cpp
M src/caffe/layers/cudnn_sigmoid_layer.cu
M src/caffe/layers/cudnn_softmax_layer.cpp
M src/caffe/layers/cudnn_softmax_layer.cu
M src/caffe/layers/cudnn_tanh_layer.cpp
M src/caffe/layers/cudnn_tanh_layer.cu
M src/caffe/layers/data_layer.cpp
M src/caffe/layers/deconv_layer.cpp
M src/caffe/layers/deconv_layer.cu
M src/caffe/layers/dropout_layer.cpp
M src/caffe/layers/dropout_layer
.cu
M src/caffe/layers/dummy_data_layer.cpp
M src/caffe/layers/eltwise_layer.cpp
M src/caffe/layers/eltwise_layer.cu
M src/caffe/layers/elu_layer.cpp
M src/caffe/layers/elu_layer.cu
M src/caffe/layers/embed_layer.cpp
M src/caffe/layers/embed_layer.cu
M src/caffe/layers/euclidean_loss_layer.cpp
M src/caffe/layers/euclidean_loss_layer.cu
M src/caffe/layers/exp_layer.cpp
M src/caffe/layers/exp_layer.cu
M src/caffe/layers/filter_layer.cpp
M src/caffe/layers/filter_layer.cu
M src/caffe/layers/flatten_layer.cpp
M src/caffe/layers/hdf5_data_layer.cpp
M src/caffe/layers/hdf5_data_layer.cu
M src/caffe/layers/hdf5_output_layer.cpp
M src/caffe/layers/hdf5_output_layer.cu
M src/caffe/layers/hinge_loss_layer.cpp
M src/caffe/layers/im2col_layer.cpp
M src/caffe/layers/im2col_layer.cu
M src/caffe/layers/image_data_layer.cpp
M src/caffe/layers/infogain_loss_layer.cpp
M src/caffe/layers/inner_product_layer.cpp
M src/caffe/layers/inner_product_layer.cu
M src/caffe/layers/input_layer.cpp
M src/caffe/layers/log_layer.cpp
M src/ca
ffe/layers/log_layer.cu
M src/caffe/layers/loss_layer.cpp
M src/caffe/layers/lrn_layer.cpp
M src/caffe/layers/lrn_layer.cu
M src/caffe/layers/lstm_layer.cpp
M src/caffe/layers/lstm_unit_layer.cpp
M src/caffe/layers/lstm_unit_layer.cu
M src/caffe/layers/memory_data_layer.cpp
M src/caffe/layers/multinomial_logistic_loss_layer.cpp
M src/caffe/layers/mvn_layer.cpp
M src/caffe/layers/mvn_layer.cu
M src/caffe/layers/neuron_layer.cpp
M src/caffe/layers/parameter_layer.cpp
M src/caffe/layers/pooling_layer.cpp
M src/caffe/layers/pooling_layer.cu
M src/caffe/layers/power_layer.cpp
M src/caffe/layers/power_layer.cu
M src/caffe/layers/prelu_layer.cpp
M src/caffe/layers/prelu_layer.cu
M src/caffe/layers/recurrent_layer.cpp
M src/caffe/layers/recurrent_layer.cu
M src/caffe/layers/reduction_layer.cpp
M src/caffe/layers/reduction_layer.cu
M src/caffe/layers/relu_layer.cpp
M src/caffe/layers/relu_layer.cu
M src/caffe/layers/reshape_layer.cpp
M src/caffe/layers/rnn_layer.cpp
M src/caffe/layers/scale_layer.cpp
M src/caffe/layer
s/scale_layer.cu
M src/caffe/layers/sigmoid_cross_entropy_loss_layer.cpp
M src/caffe/layers/sigmoid_cross_entropy_loss_layer.cu
M src/caffe/layers/sigmoid_layer.cpp
M src/caffe/layers/sigmoid_layer.cu
M src/caffe/layers/silence_layer.cpp
M src/caffe/layers/silence_layer.cu
M src/caffe/layers/slice_layer.cpp
M src/caffe/layers/slice_layer.cu
M src/caffe/layers/softmax_layer.cpp
M src/caffe/layers/softmax_layer.cu
M src/caffe/layers/softmax_loss_layer.cpp
M src/caffe/layers/softmax_loss_layer.cu
M src/caffe/layers/split_layer.cpp
M src/caffe/layers/split_layer.cu
M src/caffe/layers/spp_layer.cpp
M src/caffe/layers/swish_layer.cpp
M src/caffe/layers/swish_layer.cu
M src/caffe/layers/tanh_layer.cpp
M src/caffe/layers/tanh_layer.cu
M src/caffe/layers/threshold_layer.cpp
M src/caffe/layers/threshold_layer.cu
M src/caffe/layers/tile_layer.cpp
M src/caffe/layers/tile_layer.cu
M src/caffe/layers/window_data_layer.cpp
M src/caffe/net.cpp
M src/caffe/parallel.cpp
M src/caffe/proto/caffe.proto
M src/caffe/solver.cpp
M sr
c/caffe/solvers/adadelta_solver.cpp
M src/caffe/solvers/adadelta_solver.cu
M src/caffe/solvers/adagrad_solver.cpp
M src/caffe/solvers/adagrad_solver.cu
M src/caffe/solvers/adam_solver.cpp
M src/caffe/solvers/adam_solver.cu
M src/caffe/solvers/nesterov_solver.cpp
M src/caffe/solvers/nesterov_solver.cu
M src/caffe/solvers/rmsprop_solver.cpp
M src/caffe/solvers/rmsprop_solver.cu
M src/caffe/solvers/sgd_solver.cpp
M src/caffe/solvers/sgd_solver.cu
M src/caffe/syncedmem.cpp
M src/caffe/test/CMakeLists.txt
M src/caffe/test/test_accuracy_layer.cpp
M src/caffe/test/test_argmax_layer.cpp
M src/caffe/test/test_batch_norm_layer.cpp
M src/caffe/test/test_batch_reindex_layer.cpp
M src/caffe/test/test_benchmark.cpp
M src/caffe/test/test_bias_layer.cpp
M src/caffe/test/test_blob.cpp
M src/caffe/test/test_caffe_main.cpp
M src/caffe/test/test_common.cpp
M src/caffe/test/test_concat_layer.cpp
M src/caffe/test/test_contrastive_loss_layer.cpp
M src/caffe/test/test_convolution_layer.cpp
M src/caffe/test/test_crop_layer.cpp
M src/
HEAD is now at b5ede488 Added TX2 JetPack3.3 support
caffe/test/test_data/generate_sample_data.py
M src/caffe/test/test_data/sample_data.h5
M src/caffe/test/test_data/sample_data_2_gzip.h5
M src/caffe/test/test_data/sample_data_list.txt
M src/caffe/test/test_data/solver_data.h5
M src/caffe/test/test_data/solver_data_list.txt
M src/caffe/test/test_data_layer.cpp
M src/caffe/test/test_data_transformer.cpp
M src/caffe/test/test_db.cpp
M src/caffe/test/test_deconvolution_layer.cpp
M src/caffe/test/test_dummy_data_layer.cpp
M src/caffe/test/test_eltwise_layer.cpp
M src/caffe/test/test_embed_layer.cpp
M src/caffe/test/test_euclidean_loss_layer.cpp
M src/caffe/test/test_filler.cpp
M src/caffe/test/test_filter_layer.cpp
M src/caffe/test/test_flatten_layer.cpp
M src/caffe/test/test_gradient_based_solver.cpp
M src/caffe/test/test_hdf5_output_layer.cpp
M src/caffe/test/test_hdf5data_layer.cpp
M src/caffe/test/test_hinge_loss_layer.cpp
M src/caffe/test/test_im2col_kernel.cu
M src/caffe/test/test_im2col_layer.cpp
M src/caffe/test/test_image_data_layer.cpp
M src/caffe/test/t
est_infogain_loss_layer.cpp
M src/caffe/test/test_inner_product_layer.cpp
M src/caffe/test/test_internal_thread.cpp
M src/caffe/test/test_io.cpp
M src/caffe/test/test_layer_factory.cpp
M src/caffe/test/test_lrn_layer.cpp
M src/caffe/test/test_lstm_layer.cpp
M src/caffe/test/test_math_functions.cpp
M src/caffe/test/test_maxpool_dropout_layers.cpp
M src/caffe/test/test_memory_data_layer.cpp
M src/caffe/test/test_multinomial_logistic_loss_layer.cpp
M src/caffe/test/test_mvn_layer.cpp
M src/caffe/test/test_net.cpp
M src/caffe/test/test_neuron_layer.cpp
M src/caffe/test/test_platform.cpp
M src/caffe/test/test_pooling_layer.cpp
M src/caffe/test/test_power_layer.cpp
M src/caffe/test/test_protobuf.cpp
M src/caffe/test/test_random_number_generator.cpp
M src/caffe/test/test_reduction_layer.cpp
M src/caffe/test/test_reshape_layer.cpp
M src/caffe/test/test_rnn_layer.cpp
M src/caffe/test/test_scale_layer.cpp
M src/caffe/test/test_sigmoid_cross_entropy_loss_layer.cpp
M src/caffe/test/test_slice_layer.cpp
M src/caffe/test/t
est_softmax_layer.cpp
M src/caffe/test/test_softmax_with_loss_layer.cpp
M src/caffe/test/test_solver.cpp
M src/caffe/test/test_solver_factory.cpp
M src/caffe/test/test_split_layer.cpp
M src/caffe/test/test_spp_layer.cpp
M src/caffe/test/test_stochastic_pooling.cpp
M src/caffe/test/test_syncedmem.cpp
M src/caffe/test/test_tanh_layer.cpp
M src/caffe/test/test_threshold_layer.cpp
M src/caffe/test/test_tile_layer.cpp
M src/caffe/test/test_upgrade_proto.cpp
M src/caffe/test/test_util_blas.cpp
M src/caffe/util/benchmark.cpp
M src/caffe/util/blocking_queue.cpp
M src/caffe/util/cudnn.cpp
M src/caffe/util/db.cpp
M src/caffe/util/db_leveldb.cpp
M src/caffe/util/db_lmdb.cpp
M src/caffe/util/hdf5.cpp
M src/caffe/util/im2col.cpp
M src/caffe/util/im2col.cu
M src/caffe/util/insert_splits.cpp
M src/caffe/util/io.cpp
M src/caffe/util/math_functions.cpp
M src/caffe/util/math_functions.cu
M src/caffe/util/signal_handler.cpp
M src/caffe/util/upgrade_proto.cpp
M src/gtest/CMakeLists.txt
M src/gtest/gtest-all.cpp
M src/gtest/gtest
.h
M src/gtest/gtest_main.cc
M tools/CMakeLists.txt
M tools/caffe.cpp
M tools/compute_image_mean.cpp
M tools/convert_imageset.cpp
M tools/extra/plot_log.gnuplot.example
M tools/extract_features.cpp
M tools/upgrade_net_proto_binary.cpp
M tools/upgrade_net_proto_text.cpp
M tools/upgrade_solver_proto_text.cpp
Caffe will be built from source now.
Adding Example calibration.bin
Adding Example openpose.bin
Adding Example 1_custom_post_processing.bin
Adding Example 01_body_from_image_default.bin
Adding Example 02_whole_body_from_image_default.bin
Adding Example 03_keypoints_from_image.bin
Adding Example 04_keypoints_from_images.bin
Adding Example 05_keypoints_from_images_multi_gpu.bin
Adding Example 06_face_from_image.bin
Adding Example 07_hand_from_image.bin
Adding Example 08_heatmaps_from_image.bin
Adding Example 09_keypoints_from_heatmaps.bin
Adding Example 10_asynchronous_custom_input.bin
Adding Example 11_asynchronous_custom_output.bin
Adding Example 12_asynchronous_custom_input_output_and_datum.bin
Adding Example 13_synchronous_custom_input.bin
Adding Example 14_synchronous_custom_preprocessing.bin
Adding Example 15_synchronous_custom_postprocessing.bin
Adding Example 16_synchronous_custom_output.bin
Adding Example 17_synchronous_custom_all_and_datum.bin
Adding Example 1_thread_user_processing_function.bin
Adding Example 2_thread_user_input_processing_output_and_datum.bin
Adding Example handFromJsonTest.bin
Adding Example resizeTest.bin
Download the models.
Downloading BODY_25 model...
Model already exists.
Downloading body (COCO) model...
Model already exists.
Downloading body (MPI) model...
Model already exists.
Downloading face model...
Model already exists.
Downloading hand model...
Model already exists.
Models Downloaded.
Configuring done
Generating done
One last thing the following line worked with my current logitech webcam so I know that GST is functioning the way it should.
nvidia@nvidia-desktop:~/datacapture/openpose$ gst-launch-1.0 v4l2src device="/dev/video0" ! "video/x-raw, width=640, height=480, format=(string)YUY2" ! xvimagesink -e
Setting pipeline to PAUSED ...
Pipeline is live and does not need PREROLL ...
Setting pipeline to PLAYING ...
New clock: GstSystemClock
Camera not open.
I’m using a on-board camera, but the camera not working on openpose.
$ ./build/examples/openpose/openpose.bin
Starting OpenPose demo...
Configuring OpenPose...
Starting thread(s)...
Auto-detecting camera index... Detected and opened camera 0.
Auto-detecting all available GPUs... Detected 1 GPU(s), using 1 of them starting at GPU 0.
select timeout
select timeout
OpenCV Error: Assertion failed (total() == 0 || data != __null) in Mat, file /home/nvidia/build_opencv/opencv/modules/core/include/opencv2/core/mat.inl.hpp, line 500
Error:
/home/nvidia/build_opencv/opencv/modules/core/include/opencv2/core/mat.inl.hpp:500: error: (-215) total() == 0 || data != __null in function Mat
Coming from:
- /home/alisports/openpose/src/openpose/producer/videoCaptureReader.cpp:getRawFrame():129
- /home/alisports/openpose/src/openpose/producer/webcamReader.cpp:bufferingThread():240
terminate called after throwing an instance of 'std::runtime_error'
what():
Error:
/home/nvidia/build_opencv/opencv/modules/core/include/opencv2/core/mat.inl.hpp:500: error: (-215) total() == 0 || data != __null in function Mat
Coming from:
- /home/alisports/openpose/src/openpose/producer/videoCaptureReader.cpp:getRawFrame():129
- /home/alisports/openpose/src/openpose/producer/webcamReader.cpp:bufferingThread():240
Aborted (core dumped)
Hi al03,
Please file a new topic. Thanks