Hi, I use python, after Gst gets rtsp streaming,
- How to get the real-time frame of the video?
- After finishing net.Detect, can the results be displayed on the same viewer?
These two problems bothered me for a week,
Checked a lot of information but couldn’t solve it,
Can anyone help provide directions, thank you.
The code is similar to the following:
import jetson.inference
import jetson.utils
import cv2
import gi
gi.require_version(‘Gst’, ‘1.0’)
from gi.repository import GObject, Gst
…
…
pipeline = Gst.parse_launch(rtsp_command)
video_sink = pipeline.get_by_name(‘source’)
video_sink.props.location = rtsp_src
width = 1024
height = 768
#Show rtsp frame,It’s okay
pipeline.set_state(Gst.State.PLAYING)
#1.How to get the number of video frames(buffer)?
frame_rgba = cv2.cvtColor(frame, cv2.COLOR_BGR2RGBA)
cuda_frame = hetson.utils.cudaFromNumpy(frame_rgba)
detections = net.Detect(cuda_frame, width, height)
#2.How to integrate the results into the same viewer?