Compilation Issue when cmake

[irwanmazlin]

Description
i tried to use your project because it has 3D detection features, but i ran into compilation issue. I would really appreciate if you could help if you encounter this problem before.

Step to reproduce

1. i clone the project in ros working space
2. then after clone i try to compile the project in my working space using catkin_make then i got an error

Versions
OPENVINO 2019_R3
realsense version 2.50

[gbr1]

Hi!
Did you source openvino env variables? It seems that cmake is not able to find openvino.
Let me know!

[irwanmazlin]

i have source the environment but still same error occurs, i have uninstall the openvino and fresh install openvino 2019_R3.1

[irwanmazlin]

seems like i dont install openvino 2019_R3 version, i install 2019_R3.1 using this link http://registrationcenter-download.intel.com/akdlm/irc_nas/16057/l_openvino_toolkit_p_2019.3.376.tgz

this figure shows my installation directory

its openvino_2019.3.376, not openvino_2019.3.334
i think it one of the reason why that error occurs

do you have the installer for openvino 2019_R3?

[irwanmazlin]

Hi mate, i manage to compile the project.

but i got new error.

i am using mobilenet ssd and set target to CPU

do you know how to solve this?

[gbr1]

Hi @irwanmazlin !
Nice to hear that you manage to install it. I don't know if other ssds works instead of mobilenet (this package was created for demos). I'm quite sure that you cannot run on cpu side but you can probably use gpu if your processor is supported and for sure the myriad x

[irwanmazlin]

owh i see,

can this project run multi camera? which part i need to change if i want to run multicamera realsense?

[gbr1]

You may have 2 options:

1. run multiple nodes on multiple myriad and each node subscribes to a different image topic
2. write a node that create a bigger image by merging images

[irwanmazlin]

owh i see,

good news
now i can use your project using CPU

thank you for your help

[irwanmazlin]

i install openvino 2021.4

your code support CPU actually

[irwanmazlin]

You may have 2 options:

1. run multiple nodes on multiple myriad and each node subscribes to a different image topic
2. write a node that create a bigger image by merging images

hi @gbr1 , i have the problem run multiple node on multiple target engine,

i have set each node subscribes different image topic but the problem is, when i run the detection, the detection didnt run the simultenous for each camera, they run one by one of each camera in the same topic,

for example, i subscribe camera 1 and publish detection image of camera 1 as "/123/1" and for camera 2 i publish detection image of camera2 as "/123/2".

but i camera see the detection of both camera in one topic, it keeps switch the image after doing the detection

this is my launch file

this is my CPP file

I would really appreciate if you could help to solve this issue.