CODEWITHSUNDEEP
machine-learningdeep-learningcaffepose-estimationopenpose
Santiago Moreno
Santiago Moreno

Reputation: 1

Issue installing open-pose

I've been following the installation steps for ubuntu (22.04) on official page (https://github.com/CMU-Perceptual-Computing-Lab/openpose/blob/master/doc/installation/0_index.md), and it seems to be installed. However, at the test part i'm running:

*# Ubuntu and Mac
./build/examples/openpose/openpose.bin --video examples/media/video.avi*

but it doesnt work, this is what i get

upo@upo-Stealth-15M-A11SEK:~/openpose$ ./build/examples/openpose/openpose.bin --video examples/media/video.avi
Starting OpenPose demo...
Configuring OpenPose...
Starting thread(s)...
Auto-detecting all available GPUs... Detected 1 GPU(s), using 1 of them starting at GPU 0.
F0718 11:32:00.771245 104120 syncedmem.cpp:71] Check failed: error == cudaSuccess (2 vs. 0)  out of memory
*** Check failure stack trace: ***
    @     0x7271ad411b03  google::LogMessage::Fail()
    @     0x7271ad4199d1  google::LogMessage::SendToLog()
    @     0x7271ad4117c2  google::LogMessage::Flush()
    @     0x7271ad41378f  google::LogMessageFatal::~LogMessageFatal()
    @     0x7271ac2af1ba  caffe::SyncedMemory::mutable_gpu_data()
    @     0x7271ac2e461c  caffe::CuDNNConvolutionLayer<>::Forward_gpu()
    @     0x7271ac2700c2  caffe::Net<>::ForwardFromTo()
    @     0x7271ad01fac7  op::NetCaffe::forwardPass()
    @     0x7271ad03b132  op::PoseExtractorCaffe::forwardPass()
    @     0x7271ad035a1c  op::PoseExtractor::forwardPass()
    @     0x7271ad033690  op::WPoseExtractor<>::work()
    @     0x7271ad065e6f  op::Worker<>::checkAndWork()
    @     0x7271ad066033  op::SubThread<>::workTWorkers()
    @     0x7271ad0732cd  op::SubThreadQueueInOut<>::work()
    @     0x7271ad069cb1  op::Thread<>::threadFunction()
    @     0x7271acadc253  (unknown)
    @     0x7271ac694ac3  (unknown)
    @     0x7271ac726850  (unknown)
    @              (nil)  (unknown)
Aborted (core dumped)

I think, i should be able to run the example because i have a RTX 2060 with 6GB but it says that i'm running out of memory... so i don't know how to solve this problem.

I just want to get the json file that open pose can provide so i can use it for a NN. if there is not a solution, is it possible to get a json file with the same format as openpose with a different tool? (as mediapipe or yolo-pose) I tried on yolo (it works properly) but the format is not the same and the network is trained to get the openpose format.

pd: i have already installed CUDA 12.4 and CUDNN 8.9 and it is suposed to be working properly as they are recognized by tensorflow

Upvotes: 0

Views: 125

Answers (0)

Related Questions