gst-yoloplugin build error (Solved)

I have been trying to use Deepstream-plugin for tiny-yolo in Jetson-TX1, but facing this error while building the Makefile of gstyoloplugin,

/usr/bin/ld: cannot find -lgstnvquery
/usr/bin/ld: cannot find -lgstnvdsmeta
collect2: error: ld returned 1 exit status
Makefile:64: recipe for target 'libgstnvyolo.so' failed

I have installed gstreamer 1.0 and have installed deep-stream-2.0 according to the manual.

Seems the linker failed to find libgstnvquery.so and libgstnvdsmeta.so.
You may try to find where these libs are installed on your system with:

sudo find / -name libgstnvquery.so
sudo find / -name libgstnvdsmeta.so

and add their directories’paths to LD_LIBRARY_PATH:

export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:<path_to_libgstnvquery.so_directory>:<path_to_libgstnvdsmeta.so_directory>

Getting this error after adding the LD_LIBRARY_PATH

/usr/bin/ld: skipping incompatible /usr/local/deepstream/libgstnvquery.so when searching for -lgstnvquery
/usr/bin/ld: cannot find -lgstnvquery
/usr/bin/ld: skipping incompatible /usr/local/deepstream/libgstnvdsmeta.so when searching for -lgstnvdsmeta
/usr/bin/ld: cannot find -lgstnvdsmeta
collect2: error: ld returned 1 exit status
Makefile:64: recipe for target 'libgstnvyolo.so' failed
make: *** [libgstnvyolo.so] Error 1

I am using deepstream -2.0 for Tesla as i need to run the TRT-Yolo app in Jetson TX-1

Hi,

For Jetson user, please use the standalone TRT app.
gst-yoloplugin is only available for Tesla user due to some DeepStream dependency.

You can find this information on the GitHub:
https://github.com/vat-nvidia/deepstream-plugins
[i]--------------------------------------------------
Note

Tegra users who are currently using Deepstream 1.5, please use the standalone TRT app as your starting point and incorporate that inference pipeline in your inference plugin.
--------------------------------------------------[/i]

Thanks.

In that app there are no detections, after running it. What should i do? And what kind of image is a calibrartion image?

@subhasis.chakraborty Can you please set kSAVE_DETECTIONS = true in network_config.cpp, do a clean build and install, run the app and then check your detections folder. You should now be able to see the detections on your test images.

Calibration images are used to generate TensorRT engines for INT8 mode.

@subhasis.chakraborty The repo has now been updated with support for running yoloplugin on tegra with Deepstream 1.5