Hi,
I try to make it work the segnet class in the jetson-inference module, with my camera /dev/video0, I have no worries, but when I try with videos, the error message appeared “segnet : failed to capture video frame” . My command being, ./segnet --network=fcn-resnet18-mhp-512x320 images/test2.mp4, and I have the similar problem when I entry an rtsp adress.
Thank you.
T.
Hi,
Could you try below command:
./segnet --network=fcn-resnet18-mhp-512x320 file://[path/to/your/mp4]
# For example
./segnet --network=fcn-resnet18-mhp-512x320 file:///home/nvidia/jetson-inference/images/test2.mp4
Thanks.
Thank you for your help.
The problem is not resolved.
I have entry the same line you advice me ./segnet --network=fcn-resnet18-mhp-512x320 file:///home/jetson-inference/build/aarch64/bin/images/test2.mp4
Now the error message is :
[gstreamer] gstDecoder -- couldn't find file '/home/jetson-inference/build/aarch64/bin/images/test2.mp4'
[gstreamer] gstDecoder -- failed to create decoder for file:///home/jetson-inference/build/aarch64/bin/images/test2.mp4
segnet: failed to create input stream
T.
Hi,
We can read a video successfully with below command:
$ ./segnet --network=fcn-resnet18-mhp-512x320 file:///opt/nvidia/deepstream/deepstream-5.1/samples/streams/sample_720p.mp4
Could you try to validate if the file do exist first?
[gstreamer] gstDecoder – couldn’t find file ‘/home/jetson-inference/build/aarch64/bin/images/test2.mp4’
Thanks.