Detection Inference

I am running the my-detection.py program (with my Jetson Nano 2GB configured in NV Power Mode: 5W) with a Raspberry PI Camera IMX219 (set to csi://0), 64GB SD Card, and using Detectnet “ssd-mobilenet-v2”, threshold=0.5 .
I am using a raspberry PI Camera IMX219.

  1. The program rarely works, most times I a [gstreamer] gstCamera::Capture () – a timeout occurs waiting for the next image buffer and reports 0 fps.

  2. When it does display objects, the image is inverted 180 degrees and only recognizes humans with some
    certainty of success other objects such as a stapler is recognized as a suitcase as an example

  3. How do I change gstCamera characteristics such as: width, height, fps, flip method and store the streaming
    video?

[gstreamer] gstCamera successfully created device csi://0
[0;32m[video] created gstCamera from csi://0
[0m------------------------------------------------
gstCamera video options:

gstCamera video options

– URI: csi://0
- protocol: csi
- location: 0
– deviceType: csi
– ioType: input
– width: 1280
– height: 720
– frameRate: 30
– numBuffers: 4
– zeroCopy: true
– flipMethod: rotate-180

hello rjaffeco,

please check your sensor stream stability.
you may see-also developer guide, Applications Using V4L2 IOCTL Directly.
or, please running with below gst pipeline, which has preview disabled and shows frame-rate only.
for example,
$ gst-launch-1.0 nvarguscamerasrc sensor-id=0 ! 'video/x-raw(memory:NVMM),width=1920, height=1080, framerate=30/1, format=NV12' ! nvvidconv ! 'video/x-raw(memory:NVMM),format=I420' ! fpsdisplaysink text-overlay=0 video-sink=fakesink sync=0 -v

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.