How to get yolov5 bounding boxes (x, y, w, h) while using deepstream-app(deepstream6.1)

Accelerated Computing Intelligent Video Analytics DeepStream SDK
1

While using DeepStream 6.1 with the YOLOv5 neural network architecture, I’d like to know if it’s possible to synchronously output the bounding boxes (x, y, w, h) for each frame in the video stream.

3

You can refer to our source code:opt\nvidia\deepstream\deepstream\sources\apps\sample_apps\deepstream-app\deepstream_app.c.

static void
write_kitti_output (AppCtx * appCtx, NvDsBatchMeta * batch_meta){
...
      fprintf (bbox_params_dump_file,
          "%s 0.0 0 0.0 %f %f %f %f 0.0 0.0 0.0 0.0 0.0 0.0 0.0 %f\n",
          obj->obj_label, left, top, right, bottom, confidence);
...
}
4

thanks for your reply.
I’m confused that how can I use this function in my config.txt.
If there a Deep Stream-wiki that I can refer to?

5

You can refer to the link below: DS_ref_app_deepstream. Because our deepstream-app demo is completely open source, it can be viewed together with the source code. opt\nvidia\deepstream\deepstream\sources\apps\sample_apps\deepstream-app

6

Thanks，that’s what I need.