Software Stack
DepthAI

ON THIS PAGE

  • Holistic Replay
  • Setup
  • Pipeline
  • Source code

Holistic Replay

This example demonstrates how to replay previously recorded holistic data, feeding multiple synchronized streams such as video and IMU back into a DepthAI pipeline for testing and development.

Setup

This example requires the DepthAI v3 API, see installation instructions.

Pipeline

Source code

Python
C++

Python

Python
GitHub
1#!/usr/bin/env python3
2
3import cv2
4import depthai as dai
5import argparse
6
7parser = argparse.ArgumentParser()
8parser.add_argument("-s", "--source", default="recordings/recording.tar", help="Recording path")
9args = parser.parse_args()
10
11# Create pipeline
12with dai.Pipeline(True) as pipeline:
13    # Define source and output
14    camRgb = pipeline.create(dai.node.Camera).build(dai.CameraBoardSocket.CAM_A)
15    camRgbOut = camRgb.requestOutput((1920, 1080), fps = 30)
16
17    imu = pipeline.create(dai.node.IMU)
18    imu.enableIMUSensor(dai.IMUSensor.ACCELEROMETER_RAW, 500);
19    imu.enableIMUSensor(dai.IMUSensor.GYROSCOPE_RAW, 400);
20    imu.setBatchReportThreshold(100)
21
22    pipeline.enableHolisticReplay(args.source)
23
24    videoQueue = camRgbOut.createOutputQueue()
25    imuQueue = imu.out.createOutputQueue()
26
27    # Connect to device and start pipeline
28    pipeline.start()
29    while pipeline.isRunning():
30        videoIn : dai.ImgFrame = videoQueue.get()
31        imuData : dai.IMUData = imuQueue.get()
32
33        # Get BGR frame from NV12 encoded video frame to show with opencv
34        # Visualizing the frame on slower hosts might have overhead
35        cv2.imshow("video", videoIn.getCvFrame())
36
37        for packet in imuData.packets:
38            print(f"IMU Accelerometer: {packet.acceleroMeter}")
39            print(f"IMU Gyroscope: {packet.gyroscope}")
40
41        if cv2.waitKey(1) == ord('q'):
42            break

Need assistance?

Head over to Discussion Forum for technical support or any other questions you might have.