Add SDK to a project

The Live Sense SDK for Linux is provided as a ready-to-use Linux Debian package with few example projects.

Obtain the Live Sense SDK for Linux

To obtain the Live Sense SDK for Linux, contact your HERE Account representative.

There are three variants of the Live Sense SDK for Linux:

  1. TensorRT provided as livesense-sdk-Linux_tensorrt.deb
  2. TensorFlow Lite provided as livesense-sdk-Linux_tflite.deb
  3. Coral Edge TPU provided as livesense-sdk-Linux_edge_tpu.deb

To install the Live Sense SDK for Linux, run one of the following commands depending on the selected package:

  • For TensorRT package:

    sudo apt install ./livesense-sdk-Linux_tensorrt.deb
  • For TensorFlow Lite package:

    sudo apt install ./livesense-sdk-Linux_tflite.deb
  • For Coral Edge TPU package:

    sudo apt install ./livesense-sdk-Linux_edgetpu.deb


There might be some version mismatch for the dependencies like opencv, openssl or TensorRT libraries when you try to include the SDK in your application. To rectify, you can create symlinks from the library existing in your system to the required versions for Live Sense SDK for Linux. For example:

sudo ln -s /usr/lib/aarch64-linux-gnu/ /usr/lib/aarch64-linux-gnu/
sudo ln -s /usr/lib/aarch64-linux-gnu/ /usr/lib/aarch64-linux-gnu/
sudo ln -s /usr/lib/aarch64-linux-gnu/ /usr/lib/aarch64-linux-gnu/

In the example above, *.so.6 are the libraries used to build the Debian package and *.so.5.1.6 could be those installed in the target system. Be aware that using symlinks might cause certain risks like application crashes, data corruption etc.

Add to project via CMake

    # Core
    # ML model classes
    # ML runtime of installed SDK build
    # livesense_tensorrt
    # Trackers
    # Observation library
    # Heuristics (deviation, etc.)
    # Camera

add_executable(YOUR_APP_NAME YOUR_APP_NAME.cpp)

target_link_libraries(YOUR_APP_NAME PUBLIC ${liveSenseLibraries})

results matching ""

    No results matching ""