Depth and Color frame data read

 config.enable_stream(rs.stream.color, rs.format.bgr8, 30)
config.enable_stream(rs.stream.depth, rs.format.z16, 30) 

Traceback (most recent call last):
  File "read_bag.py", line 40, in <module>
    pipeline.start(config)
RuntimeError: Couldn't resolve requests

• Aznie Syaarriehaah

  • May 12, 2021 06:42

  Hi Charmipatel17071995,
  From your snippet code, you are missing the resolution for the FPS of the stream. Have a look at this opencv_viewer_example.py line 22 for the code or else you also can pass and export the argument as done in this export_ply_example.py script.
  
  Regards,
  Aznie
  Intel RealSense Customer Support

  0

  Comment actions Permalink
• Charmipatel17071995

  • May 12, 2021 07:49

  Hello, 

  Thanks for your quick help. I tried using the resolution, but getting the same error.

  config.enable_stream(rs.stream.color, 960, 540, rs.format.bgr8, 30)
  config.enable_stream(rs.stream.depth, 640, 480, rs.format.z16, 30)

  0

  Comment actions Permalink
• Aznie Syaarriehaah

  • May 12, 2021 08:18

  Hi Charmipatel17071995,
  Did you get the recorded .bag file from RealSense Viewer? Please provide us more detail on the python programs you are running from the examples directory? Also, can you try to connect your camera with RealSense Viewer and check the USB detected at the top of the viewer? Please shares the screenshot with us.
  
  Regards,
  Aznie
  Intel RealSense Customer Support

  0

  Comment actions Permalink
• Charmipatel17071995

  • May 12, 2021 10:54
  • Edited

  Hello,

  Yes I have recorded *.bag file from the viewer. I am trying to write a python code which extracts depth and color frame from recorded *.bag file and display it using opencv window. The camera is connected perfectly and I view the parameters in the rs-rosbag-inspector properly.

  pipeline = rs.pipeline()
  
  # Create a config object
  config = rs.config()
  
  rs.config.enable_device_from_file(config, '*.bag', False)
  config.enable_stream(rs.stream.color)
  config.enable_stream(rs.stream.depth)
  
  pipeline.start(config)
  
  while True:
         # Get frameset of depth
         frames = pipeline.wait_for_frames()
  
         # Get depth frame
         depth_frame = frames.get_depth_frame()
  
         # Get color frame
         color_frame = frames.get_color_frame()

   

  I tried simply doing this as per your suggestion and it works fine.

  config.enable_stream(rs.stream.color)
  config.enable_stream(rs.stream.depth)

   

  I also want to extract the confidence value of each pixel in a frame. Is it possible to get that with L515?

  Regards,

  Charmi

  0

  Comment actions Permalink
• Aznie Syaarriehaah

  • May 14, 2021 07:29

  Hi Charmipatel17071995,
  Yes, you can access the confidence value within L515, Here is the example of how to enable the stream confidence for each sensor:
  config.enable_stream(rs2_stream::RS2_STREAM_CONFIDENCE, 640, 480, RS2_FORMAT_RAW8, 30);
  
  Regards,
  Aznie
  Intel RealSense Customer Support

  0

  Comment actions Permalink
• Charmipatel17071995

  • May 17, 2021 07:58

  Hello,

  I have recorded the stream after enabling the stream confidence. But I do not know how to access the stream after enabling it. Can you please help me with that?

  Regards,

  Charmi 

  0

  Comment actions Permalink
• Aznie Syaarriehaah

  • May 19, 2021 05:30

  Hi Charmipatel,
  You can access the stream by doing this:

  confidence_frame = frame_data.first_or_default(RS2_STREAM_CONFIDENCE);

  Regards,
  Aznie
  Intel RealSense Customer Support

  0

  Comment actions Permalink
• Charmipatel17071995

  • May 19, 2021 05:58

  Hi,

  Thank you very much. I am able to access the confidence stream after converting the stream to numpy array. I found that the confidence maps are stored in the following format. I want to extract the data in the following format for each pixel.

  

   

  confidence_output = confidence_frame.get_data()
  confidence = np.asanyarray(confidence_output)

   

  Regards,

  Charmi

  0

  Comment actions Permalink

Please sign in to leave a comment.