CMU-Perceptual-Computing-Lab / openpose

OpenPose: Real-time multi-person keypoint detection library for body, face, hands, and foot estimation
https://cmu-perceptual-computing-lab.github.io/openpose
Other
30.68k stars 7.82k forks source link

Reduce incoming FPS + Image resolution of FLIR camera #1516

Closed sebo361 closed 4 years ago

sebo361 commented 4 years ago

Issue Summary

In general, when running the system with multiple FLIR cameras at the same time, the system crashes due to bandwith issue. I found a way of solving this problem as I either reduce the incoming frame rate or image resolution (reduced it from 1240x1048 to 640x480) or reducing both. Now I am facing the same problem when running OpenPose with more than one Flir camera connected as the maximum FPS and image resolution is received. Hence, I would like to lower the received FPS/image resolution of Flir camera stream. The proposed flag "--fps_max" reduces the displayed video stream only and does not handle the incoming video data.

Is there any solution to reduce the incoming FPS and/or image resolution of Flir cameras? How do I change the code (although I would not like to compile OpenPose again) in order to set FPS/image resolution?

Executed Command (if any)

./build/examples/openpose/openpose.bin --num_gpu 0 --flir_camera

OR

./build/examples/openpose/openpose.bin --num_gpu 0 --flir_camera --frame_undistort --write_images ~/Desktop/extrinsics

Type of Issue

You might select multiple topics, delete the rest:

Your System Configuration

CONFIGURING TRIGGER Configuring trigger... Trigger mode disabled... Trigger mode turned back on... CONFIGURING TRIGGER Configuring trigger... Trigger mode disabled... Trigger mode turned back on... Camera 0 acquisition mode set to continuous... Choosing maximum resolution for flir camera (1280 x 1024). Camera 0 started acquiring images... Camera 1 acquisition mode set to continuous... Choosing maximum resolution for flir camera (1280 x 1024). Camera 1 started acquiring images...

Reading (and sorting by) serial numbers... Camera 0 serial number set to 19357407... Camera 1 serial number set to 19357408... Image incomplete with image status 3... In /home/lindera/openpose/src/openpose/producer/spinnakerWrapper.cpp:bufferingThread():491 Image incomplete with image status 9... In /home/lindera/openpose/src/openpose/producer/spinnakerWrapper.cpp:bufferingThread():491 Image incomplete with image status 3... In /home/lindera/openpose/src/openpose/producer/spinnakerWrapper.cpp:bufferingThread():491 Image incomplete with image status 3... In /home/lindera/openpose/src/openpose/producer/spinnakerWrapper.cpp:bufferingThread():491 Image incomplete with image status 3... In /home/lindera/openpose/src/openpose/producer/spinnakerWrapper.cpp:bufferingThread():491

  1. OpenPose version: Latest GitHub code

  2. General configuration:

    • Installation mode: CMake, sh script, manual Makefile installation

    • Operating system (lsb_release -a in Ubuntu): Distributor ID: Ubuntu, Description: Pop!_OS 18.04 LTS, Release: 18.04, Codename: bionic

    • Operating system version (e.g., Ubuntu 16, Windows 10, ...): Ubuntu 18.4

    • Release or Debug mode? (by default: release):

    • Compiler (gcc --version in Ubuntu or VS version in Windows): gcc (Ubuntu 7.4.0-1ubuntu1~18.04.1) 7.4.0

  3. Non-default settings:

    • 3-D Reconstruction module added? (by default: no): YES
    • Any other custom CMake configuration with respect to the default version? (by default: no): NO
  4. 3rd-party software:

    • Caffe version: Default from OpenPose
    • CMake version (cmake --version in Ubuntu): cmake version 3.13.4
    • OpenCV version: pre-compiled apt-get install libopencv-dev
  5. If GPU mode issue:

    • CUDA version (cat /usr/local/cuda/version.txt in most cases): 10.0.
    • cuDNN version: 7.5.1
    • GPU model (nvidia-smi in Ubuntu): GeForce GTX 107
sebo361 commented 4 years ago

found it with using the flags --fps_max and --camera_resolution :)