Hutch07

  • 2 Jan
  • Joined Jun 21, 2023
  • 0 best answers
  • In X_Link_device_Not_Found
  • In X_Link_device_Not_Found

    I have 2 cameras that were running fine. There were no changes to the code and now have this error

    Exception has occurred: RuntimeError

    • Failed to find device after booting, error message: X_LINK_DEVICE_NOT_FOUND

    I have been getting a warning about DeprecationWarning: Use constructor taking 'UsbSpeed' instead

    device = stack.enter_context(dai.Device(openvino_version, device_info, usb2_mode))

    I do not see information on how to use UsbSpeed. m Do I set it to "High"?

    • In Use the OAK-D-Pro IMU to determine velocity.

      This will take a bit. I need approval…

      • In Use the OAK-D-Pro IMU to determine velocity.
      • In Sync 2 Oak Pro Cameras, set the iso/exposure and record.

        That absolutely worked. I updated my def create_pipeline(cam_list) in the file and took out the "cam" loop to make it a bit simpler.

        Thank You Jaka!

        Hutch

      • In Sync 2 Oak Pro Cameras, set the iso/exposure and record.

        This code starts 2 pipelines for 2 OAK-D cameras but sending the manual exposure has no effect on the cameras. They stay in auto-exposure mode.

        • In Sync 2 Oak Pro Cameras, set the iso/exposure and record.

          When I use the depthai_sdk, I do not have a problem setting the exposure and can save the video.

          When I use depthai like in "luxonis/depthai-experiments/blob/master/gen2-syncing/host-multiple-OAK-sync.py"

          I can sync the images but setting the exposure seems to do nothing.

          The Mono Camera exposure control example works with a single camera. "https://docs.luxonis.com/projects/api/en/latest/samples/MonoCamera/mono_camera_control/#mono-camera-control"

        • In Sync 2 Oak Pro Cameras, set the iso/exposure and record.

          If anyone can give me a hint on how to set the exposure with a pipeline, I think the modifications to host-multiple-OAK-sync.py would work. The motion blur I am getting will not work with auto exposure.

          • In Sync 2 Oak Pro Cameras, set the iso/exposure and record.

            "luxonis/depthai-experiments/blob/master/gen2-syncing/host-multiple-OAK-sync.py"

            works fine. I will save each frame as a file with frame-count in the name. Then I can save a CSV file with frame-count, IMU data, timestamps. I think somehow saving video would be faster.

            :No MCAP will not work. Only VIDEO

          • In Does MCAP recording work any longer on OAK-D with depthai?

            jakaskerl ROSAG also fails.

            Exception has occurred: ModuleNotFoundError

            No module named 'rosbags'

            File "D:\Programming\OAK_D\M_exp_rec.py", line 40, in <module> oakL.record([ left.out.encoded, right.out.encoded], out_path, RecordType.ROSBAG) ModuleNotFoundError: No module named 'rosbags'

              • In Rosbag Replaying

                ADP I am trying to nearly the same with 2 OAK-D Pro cameras. I was attempting MCAP. I will also try ROSBAG

                • In Does MCAP recording work any longer on OAK-D with depthai?

                  I have tried several examples from the depthai and depthai-sdk (https://docs.luxonis.com/projects/sdk/en/latest/samples/recording/SDK_mcaprecord/) and if the RecordType is MCAP or ROSBAG they fail with Module Not Found Error no module mcap_ros1. They will work if the RecordType is VIDEO. I am trying to record IMU data and I am not sure if the IMU data will be encoded in the VIDEO format.

                  I am using:

                  • depthai-sdk 1.13.1
                  • depthai 2.22.0.0
                  • Windows 10
                  • In Sync 2 Oak Pro Cameras, set the iso/exposure and record.
                  • In Sync 2 Oak Pro Cameras, set the iso/exposure and record.
                  • In Sync 2 Oak Pro Cameras, set the iso/exposure and record.

                    I would like to synchronize my 2 Oak-D Pros and record the video (CAMB and CAM_C only) with the IMU information for each frame (15 FPS) exported to a CSV. This was going well until I added adjusting the exposure of each camera. I had to upgrade the depthai-SDK to 1.13 for the exposure and now the reworked example code for record.py needs a major overhaul. It was written for depthai-sdk 1.2.

                    Can you suggest psudo code (or actual) on how to record these synchronized 2 streams with IMU info?

                    • jakaskerl replied to this.

                    • Hi Hutch07
                      You are missing the link between Input (XLINK) and the control.

                      controlLeft = pipeline.create(dai.node.XLinkIn) #controlIn
controlLeft.setStreamName('controlLeft')
controlRight = pipeline.create(dai.node.XLinkIn) #controlIn
controlRight.setStreamName('controlRight')

controlLeft.out.link(cam[c].inputControl)
controlRight.out.link(cam[c].inputControl)

                      Thanks,
                      Jaka

                      • In Independent ISO Controls With Manual Exposure

                        Have you looked at the camera_control.py in the SDK?

                      • In Record video output with annotations

                        jakaskerl I was not closing the cameras so the encoded files were missing information.

                        Thanks.

                        • In How to decode video recorded with encode='H265'

                          Thanks jakaskerl I was also trying replay but it turns out that I was not closing my cameras in the code I was using to record the video. I changed the recording code to this:

                          Record_With_Exposure

                          Thanks again for your help.

                          • In How to decode video recorded with encode='H265'

                            My code I am using to record the video is :

                            left = oak.create_camera('left', resolution='720p', fps=20, encode='H265')
right = oak.create_camera('right', resolution='720p', fps=20, encode='H265')

# Synchronize & save all (encoded) streams
oakR =OakCamera("1844301081B8AC0F00")
leftR = oakR.create_camera('left', resolution='720p', fps=20, encode='H265')
rightR = oakR.create_camera('right', resolution='720p', fps=20, encode='H265')

oak.record([ left.out.encoded, right.out.encoded], out_path, RecordType.VIDEO)
oakR.record([ leftR.out.encoded, rightR.out.encoded], out_path, RecordType.VIDEO)`

                            The code I am trying to use to open the mp4 file is:

                             from depthai_sdk import OakCamera

with OakCamera(recording='E:\\Junk\\vid\\1-1844301081B8AC0F00\\CAM_B_bitstream.mp4') as oak:

                                left =oak.create_camera('left', resolution='720p', fps=20, encode='H265')

                                oak.visualize([left.out.camera], scale=2/3, fps=True)

                                oak.visualize(left.out, fps=True)

                                oak.start(blocking=True)

                            My error is this :

                            Exception has occurred: TypeError
                            OakCamera.init() got an unexpected keyword argument 'recording'
                            File "D:\Programming\OAK_D\replaysdk.py", line 2, in <module>
                            with OakCamera(recording='E:\Junk\vid\1-1844301081B8AC0F00\CAM_B_bitstream.mp4') as oak:
                            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                            TypeError: OakCamera.init() got an unexpected keyword argument 'recording'

                            • In Record video output with annotations

                              How can I view or decode an mp4 video? I have the camera recording OK but have not found anything that can open it.

                              from depthai_sdk import OakCamera, RecordType

                              from multiprocessing.sharedctypes import Value

                              import depthai as dai

                              import contextlib

                              import math

                              import time, datetime

                              from pathlib import Path

                              import signal

                              import threading

                              import csv

                              out_path = r"E:\Junk\vid"

                              OUTPUT_DIR = Path(r"F:\CAMERAS")

                              Output_csv = Path.joinpath(OUTPUT_DIR, "CSV")

                              print("Output", Output_csv)

                               

                              def write_times(outime, big_array):

                                  print("Writing Timestamp CSV.")

                                  outname = outime.replace(":", "_") + ".csv"

                                  outfile = Path.joinpath(Output_csv, outname)

                                  header = ["Device", "Cam", "Timestamp"]

                                  with open(outfile,"w", newline='') as csvfile:

                                      spamwriter = csv.writer(csvfile, delimiter=',')

                                      spamwriter.writerow(header)

                                      for rowi in big_array:

                                          spamwriter.writerow(rowi)

                                      return True

                               

                              with OakCamera() as oak:

                                  left = oak.create_camera('left', resolution='800p', fps=20, encode='H265')

                                  right = oak.create_camera('right', resolution='800p', fps=20, encode='H265')

                               

                                  # Synchronize & save all (encoded) streams

                                  oak.record([ left.out.encoded, right.out.encoded], out_path, RecordType.VIDEO)

                                  # Show left stream

                                  oak.visualize([left.out.camera], scale=2/3, fps=True)

                                

                               

                                  oak.start()

                                  left.control.set_exposure_iso(1500, 600)

                                  while oak.running():

                                 

                                      key = oak.poll()

                                      if key == ord('i'):

                                          left.control.exposure_time_down()

                                      elif key == ord('o'):

                                          left.control.exposure_time_up()

                                      elif key == ord('k'):

                                          left.control.sensitivity_down()

                                      elif key == ord('l'):

                                          left.control.sensitivity_up()

                               

                                      elif key == ord('e'): # Switch to auto exposure

                                          left.control.send_controls({'exposure': {'auto': True}}

                              [upl-image-preview url=https://discuss.luxonis.com/assets/files/2023-09-09/1694286258-701158-image.png]