- Notifications
You must be signed in to change notification settings - Fork1
A Python wrapper for Azure-Kinect-Sensor-SDK
License
NotificationsYou must be signed in to change notification settings
ykoneee/k4a-python
Folders and files
| Name | Name | Last commit message | Last commit date | |
|---|---|---|---|---|
Repository files navigation
Since my Azure-Kinect is broken, it will not be updated in the future.You can get the project source code from pypi.
This library is a wrapper in Python for Azure-Kinect-Sensor-SDK
- depth\color\ir image numpy data
- depth (colorize)
- depth to color transform
- IMU data
- read camera setting
- set camera setting
- some trivial functions, such as get_installed_count
- installAzure Kinect SDK first
- install pybind11 by pip,
pip install pybind11
pip install k4a-pythonpip install only have test on ubuntu 18.04, but it should aslo work well with other linux distributions.
pip on windows has not been tested, will test as soon.
- other official c++ API
- other official modules, like k4arecord
- official example
- Microphone data
The Python API mainly refers to theofficial C + + API
(Will have better example)
dirty example code :
import pyk4aimport cv2import numpy as npprint(pyk4a.Device.get_installed_count())device=pyk4a.Device.open(pyk4a.K4A_DEVICE_DEFAULT)print(device.get_serialnum())config = pyk4a.Configuration()config.depth_mode = pyk4a.K4A_DEPTH_MODE_WFOV_2X2BINNEDconfig.color_resolution = pyk4a.K4A_COLOR_RESOLUTION_720Pconfig.camera_fps = pyk4a.K4A_FRAMES_PER_SECOND_30config.color_format = pyk4a.K4A_IMAGE_FORMAT_COLOR_BGRA32config.synchronized_images_only = Truecalibration = device.get_calibration(config.depth_mode, config.color_resolution)transformation = pyk4a.Transformation(calibration)device.start_cameras(config)device.start_imu()capture=pyk4a.Capture()a=10while a>0: a-=1 print(a) if device.get_capture(capture,1000): img=capture.get_color_image() # img=transformation.depth_image_to_color_camera(img) # print(img.get_device_timestamp()) print(device.get_raw_calibration()) img=img.numpy() # state=device.get_imu_sample() # print(state) cv2.imshow('t',img) cv2.waitKey(1)