Defining the Future of 3D Machine Vision
Project description
Zivid Python
Zivid Python is the official Python package for Zivid 3D cameras. Read more about Zivid at zivid.com.
Contents: Installation | Quick Start | Examples | Versioning | License | Support | Test Matrix
Installation
Dependencies
- Python version 3.6 or higher
- Zivid SDK version 2.9.0 (see here for help)
- Compiler with C++17 support
Ubuntu users must install Python headers (apt install python3-dev
) in addition to the regular python3
package.
Windows users also needs to make sure that the Zivid SDK installation folder is in system PATH
before using the package, not only the terminal PATH variable. The default install location that should be added to system PATH
is C:\Program Files\Zivid\bin
.
Installing official version from PyPI using PIP
After having installed the latest Zivid SDK, the easiest way to install Zivid Python is to use PIP to fetch the latest official version from PyPI:
pip install zivid
Note:
If you don't use the latest Zivid SDK version you need to manually specify the version. See Versioning.
Installation may take some time since the setup.py
script will download additional dependencies and compile C++ source code in the background.
On some systems Python 3 pip
is called pip3
. In this guide we assume it is called pip
. When using PIP version 19 or higher build dependencies are handled automatically.
Old PIP
If you are using a version of PIP older than version 19 please manually install the dependencies listed in pyproject.toml before installing zivid.
pip install <packages listed in pyproject.toml>
pip install zivid
Installing from source
git clone <zivid-python clone URL>
cd zivid-python
pip install .
You may want to build Zivid Python against a different (but compatible) version of Zivid SDK. An example would be if Zivid SDK 2.1 was released but the official
Zivid Python still formally only supports SDK 2.0. Since all the features of the 2.0 API exist in the 2.1 API, Zivid Python can still be built with the new SDK
(but without wrapping the latest features). In order to achieve this, edit setup.py
to target the new SDK version before doing pip install .
. Note that
this option is considered experimental/unofficial.
Quick Start
Point cloud capture
To quickly capture a point cloud using default settings, run the following code:
import zivid
app = zivid.Application()
camera = app.connect_camera()
settings = zivid.Settings(acquisitions=[zivid.Settings.Acquisition()])
frame = camera.capture(settings)
frame.save("result.zdf")
Instead of using the API to define capture settings, it is also possible to load them from YML files that have been exported from Zivid Studio or downloaded from the Zivid Knowledge Base settings library. This can be done by providing the filesystem path to such a file, for example:
settings = Settings.load("ZividTwo_Settings_2xHDR_Normal.yml")
frame = camera.capture(settings)
Point cloud data access
Data can easily be accessed in the form of Numpy arrays:
import zivid
app = zivid.Application()
camera = app.connect_camera()
settings = zivid.Settings(acquisitions=[zivid.Settings.Acquisition()])
frame = camera.capture(settings)
xyz = frame.point_cloud().copy_data("xyz") # Get point coordinates as [Height,Width,3] float array
rgba = frame.point_cloud().copy_data("rgba") # Get point colors as [Height,Width,4] uint8 array
bgra = frame.point_cloud().copy_data("bgra") # Get point colors as [Height,Width,4] uint8 array
Capture Assistant
Instead of manually adjusting settings, the Capture Assistant may be used to find the optimal settings for your scene:
import zivid
app = zivid.Application()
camera = app.connect_camera()
capture_assistant_params = zivid.capture_assistant.SuggestSettingsParameters()
settings = zivid.capture_assistant.suggest_settings(camera, capture_assistant_params)
frame = camera.capture(settings)
frame.save("result.zdf")
Using camera emulation
If you do not have a camera, you can use the FileCameraZividOne.zfc
file in ZividSampleData2.zip to emulate a camera.
import zivid
app = zivid.Application()
camera = app.create_file_camera("path/to/FileCameraZividOne.zfc")
settings = zivid.Settings(acquisitions=[zivid.Settings.Acquisition()])
frame = camera.capture(settings)
frame.save("result.zdf")
Examples
Basic example programs can be found in the samples directory. Many more advanced example programs may be found in the separate zivid-python-samples repository.
Versioning
This python module is using PEP 440 for versioning. The features available in the module depends on the Zivid SDK version used when building the module. When updating this Python package it is recommended to also update to the latest Zivid SDK. Refer to the Test Matrix for supported version.
The version number of the Zivid Python module consists of six numbers. The three first numbers of the version is the semantic version of the code in this repository. The last three numbers is the version of the underlying Zivid SDK library used by the Python module.
To check which version of zivid-python that corresponds to a specific version of Zivid SDK, check out zivid-python-releases-url or run pip index versions zivid
.
Version breakdown
Zivid SDK version = 1.4.1 (semantic version)
v v v
Zivid Python module version = 1.0.0.1.4.1
^ ^ ^
Wrapper code version = 1.0.0 (semantic version)
PyPI
When installing using PIP it is possible to specify the required version. This can be useful if upgrading Zivid SDK is not desired, but you want to update Zivid Python.
Install latest version of Zivid Python using latest version of Zivid SDK
pip install zivid
Install version 1.0.0 of Zivid Python using latest version of Zivid SDK
pip install zivid==1.0.0
Install version 1.0.0 of Zivid Python using Zivid SDK version 1.4.0
pip install zivid==1.0.0.1.4.0
Install version 1.0.0 of Zivid Python using Zivid SDK version 1.3.0
pip install zivid==1.0.0.1.3.0
Support for older versions of Zivid SDK will be discontinued when they are no longer compatible with latest version of the wrapper code.
License
This project is licensed, see the LICENSE file for details. The licenses of dependencies are listed here.
Support
Please visit Zivid Knowledge Base for general information on using Zivid 3D cameras. If you cannot find a solution to your issue, please contact customersuccess@zivid.com.
Test matrix
Operating System | Python version |
---|---|
Ubuntu 23.04 | 3.11 |
Ubuntu 22.10 | 3.10 |
Ubuntu 22.04 | 3.10 |
Ubuntu 20.04 | 3.8 |
Ubuntu 18.04 | 3.6 |
Fedora 37 | 3.11 |
Fedora 36 | 3.10 |
Fedora 35 | 3.10 |
Windows 10 | 3.7, 3.8, 3.9, 3.10, 3.11 |
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distributions
Hashes for zivid-2.9.0.2.9.0-cp311-cp311-win_amd64.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 514da1187a7cf1067f9bb6e52fb328f443982715eec318cc66385d76f0f73721 |
|
MD5 | 78494ac6b6039e5382c2afe5f797b172 |
|
BLAKE2b-256 | 6b05f25b52055d3641e0fc6075cad038b8d353677ef327b3d3e0b4f00c9d824c |
Hashes for zivid-2.9.0.2.9.0-cp310-cp310-win_amd64.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 0598b326995fcb662c00d94bf02f5765cdd7a600e1cd69554610eb53f6b55021 |
|
MD5 | e42e6a4cf967169b4ad57d8b98b61e1d |
|
BLAKE2b-256 | 31cf5f44c57ca8d9230b3346364e79e88dd35c83544b8c354bf627afdf449799 |
Hashes for zivid-2.9.0.2.9.0-cp39-cp39-win_amd64.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 1180c5e2c110504c955b5391f7fc4b4b2a6e82cbb118a1dc45a82cfe7c39516c |
|
MD5 | a1e467b802048254f5a89bcc3cb10582 |
|
BLAKE2b-256 | 4e3304954f820bf676ff882fa8467a2dcebbf6a497dbe9c20d8ec642f98ba897 |
Hashes for zivid-2.9.0.2.9.0-cp38-cp38-win_amd64.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 0c166ad26a51fcf2f9a7a48e78f11af57521dd8cbc09cd8a460f3f6308c3e30d |
|
MD5 | 51a72865762f60e57e5136e0d61fb7a4 |
|
BLAKE2b-256 | 6031e218a7ea37e74fb12a0823a921eae17126e414f7de51dfd0357f977ace0b |
Hashes for zivid-2.9.0.2.9.0-cp37-cp37m-win_amd64.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | a2d6f4d63b733f90e3b6455e06e3943dcf1ed60480060777a01234c80a2752ad |
|
MD5 | 8b15ee831515c5cc87f3679ae155c810 |
|
BLAKE2b-256 | fc350d42b260c1492deeccdab5bf85f58238a329a5198e22a887825775be80b0 |