Getting Started

This page walks you through installing seekcamera-python and making your first API call against a Seek Thermal OEM camera core. By the end, you will have the library and its dependencies installed, a physical camera recognized by the SDK, and a sample application running that captures and outputs thermal frame data. Getting this baseline working first makes it straightforward to build your own capture logic on top of the same patterns.

Before you begin, make sure you have the following in place:

• Python 3.6 or later — Earlier 3.x versions may work but are not tested. Download from python.org.
• pip — Used to install the package and its dependencies.
• Seek Thermal SDK 4.x — Must be installed separately on your system before using this library. The minimum SDK runtime version for the current release (seekcamera-python 1.3.0) is 4.2.0. Obtain the SDK and its installer from thermal.com.
• Seek Thermal OEM camera core — A physical camera must be connected to your machine.
• numpy >= 1 — Installed automatically when you install the package.
• opencv-python >= 4 (optional) — Required only if you want to run the seekcamera-opencv sample application.

SDK library location

The library loader searches for the SDK shared library in these locations:

Linux (in order of precedence):

$SEEKTHERMAL_LIB_DIR/
$LD_LIBRARY_PATH/
/lib/
/lib64/
/usr/lib/
/usr/lib64/

Windows (in order of precedence):

%SEEKTHERMAL_LIB_DIR%/
C:/Program Files/Seek Thermal/Seek Thermal SDK/4.X.X/x64-windows/lib

You can set $SEEKTHERMAL_LIB_DIR (Linux) or %SEEKTHERMAL_LIB_DIR% (Windows) to point to a custom install location.

The fastest path from zero to a running thermal capture:

1. Install the Seek Thermal SDK 4.x for your platform (see thermal.com).
2. Install seekcamera-python from PyPI:
   Copy

   pip3 install seekcamera-python
   

3. Install optional dependencies for the sample applications:
   Copy

   pip3 install opencv-python
   

4. Connect your Seek Thermal OEM camera core to the machine.
5. Run the OpenCV sample to confirm everything is working:
   Copy

   python3 examples/seekcamera-opencv.py
   

If a thermal image window appears on screen, your installation is complete and the camera is recognized.

Step 1 — Install the Seek Thermal SDK

Download and run the SDK installer for your platform from thermal.com. The Python bindings are a wrapper around this SDK; the native library (libseekcamera.so on Linux or seekcamera.dll on Windows) must be present on the system before you can import the Python package.

Verify the library is in a location the loader will find (see the paths listed in Prerequisites). If you have a non-standard install path, set the appropriate environment variable:

# Linux
export SEEKTHERMAL_LIB_DIR=/path/to/sdk/lib

# Windows (Command Prompt)
set SEEKTHERMAL_LIB_DIR=C:\path\to\sdk\lib

Step 2 — Install seekcamera-python

Recommended — from PyPI:

pip3 install seekcamera-python

Alternative — from source:

git clone https://github.com/seekthermal/seekcamera-python.git
cd seekcamera-python
pip3 install -e .

The -e flag installs in editable mode, which is convenient if you plan to modify the bindings or inspect the source alongside your own code.

Success indicator: Running pip3 show seekcamera-python should display version and location information without errors.

Step 3 — Install sample application dependencies

If you want to run the bundled example applications, install their additional dependencies:

pip3 install -r requirements.examples.txt

This installs opencv-python, which is required by the seekcamera-opencv sample. The core library only requires numpy, which was already installed in Step 2.

Step 4 — Connect your camera

Physically connect your Seek Thermal OEM camera core to the machine using the appropriate interface. The SDK will detect it automatically when initialized; no additional driver installation is needed on most platforms.

Step 5 — Run a sample application

With the repository cloned (or the examples installed alongside the package), run one of the bundled samples:

# Display a live thermal image using OpenCV
python3 examples/seekcamera-opencv.py

# Export frame pixel values to a CSV file on disk
python3 examples/seekcamera-simple.py

Success indicator for seekcamera-opencv: A window opens displaying a real-time thermal image from the connected camera.

Success indicator for seekcamera-simple: A CSV file is written to the working directory containing pixel values from a captured frame.

Example 1 — Run the OpenCV display sample

This sample demonstrates connecting to a camera, receiving frames asynchronously, and rendering them to the screen. It is the fastest way to verify your setup end-to-end.

python3 examples/seekcamera-opencv.py

Expected output: An OpenCV window titled with the camera's serial number opens and displays a live, continuously updating thermal image. The terminal shows no error messages.

Example 2 — Run the CSV export sample

This sample demonstrates capturing a thermal frame and writing its pixel values to disk as a CSV file — useful for offline analysis.

python3 examples/seekcamera-simple.py

Expected output: The script connects to the camera, captures one or more frames, and writes a .csv file to the current directory. A status message in the terminal confirms the file path written.

Example 3 — Verify the import in Python

Before writing your own application, confirm the package imports correctly in your environment:

import seekcamera
print(seekcamera.__doc__)

Expected output: The module docstring is printed without any ImportError or missing-library errors. If you see an error about libseekcamera.so or seekcamera.dll not being found, revisit the SDK library path configuration in Step 1.

ImportError: cannot import name 'seekcamera'

Symptom: Running any script that imports seekcamera raises an ImportError.

Likely cause: The package was not installed in the Python environment you are currently using, or the installation failed silently.

Fix: Confirm which Python and pip you are using:

which python3
which pip3

Re-install the package into the correct environment:

pip3 install seekcamera-python

OSError or RuntimeError about missing libseekcamera.so / seekcamera.dll

Symptom: The package imports, but instantiating any SDK class raises an error indicating the native library cannot be found.

Likely cause: The Seek Thermal SDK is not installed, or its library is in a directory the loader does not search by default.

Fix: Install the Seek Thermal SDK 4.x for your platform. Then ensure the library is in a searched path or set the environment variable before running your script:

# Linux
export SEEKTHERMAL_LIB_DIR=/path/to/sdk/lib

# Windows
set SEEKTHERMAL_LIB_DIR=C:\path\to\sdk\lib

SDK version mismatch error at runtime

Symptom: The library loads, but the SDK reports a version incompatibility error when you try to connect to a camera.

Likely cause: The installed Seek Thermal SDK runtime is older than the minimum version required by your version of seekcamera-python. For version 1.3.0, the minimum SDK runtime is 4.2.0.

Fix: Update the Seek Thermal SDK to at least the minimum version shown in the compatibility table. Check your installed SDK version in the SDK's documentation or installer logs.

No camera detected / camera not found

Symptom: The sample application starts but reports no camera connected or times out waiting for a device.

Likely cause: The physical camera is not connected, is connected to an unsupported port, or the OS has not enumerated the device yet.

Fix: Disconnect and reconnect the camera, wait a few seconds, and re-run the application. Confirm the device appears in your OS device list (lsusb on Linux or Device Manager on Windows). Check that your host architecture is supported by the SDK.

ModuleNotFoundError: No module named 'cv2'

Symptom: Running seekcamera-opencv.py fails immediately with a missing cv2 module error.

Likely cause: opencv-python is not installed in the current Python environment.

Fix:

pip3 install opencv-python

Or install all sample dependencies at once:

pip3 install -r requirements.examples.txt