dnlcrl / ai-assisted-annotation-client

Client side integration example source code and libraries for AI-Assisted Annotation SDK

NVIDIA AI-Assisted Annotation Client

NVIDIA AI-Assisted Annotation SDK follows a client-server approach to integrate into an application. Clara Train SDK container on ngc.nvidia.com is generally available for download, annotation server is included in the package. Once a user has setup the AI-Assisted Annotation Server, user can use either C++ or Python client to integrate the SDK into an existing medical imaging application.

Supported Platforms

AI-Assisted Annotation is a cross-platform C++/Python Client API to communicate with AI-Assisted Annotation Server for NVIDIA and it officially supports:

• Linux (Ubuntu16+)
• macOS (High Sierra and above)
• Windows (Windows 10)

Plugins

Following plugins integrate with NVIDIA AI-Assisted Annotation through either C++/Python Client APIs

• NVIDIA MITK Plugin (based on c++ client APIs)
• NVIDIA 3D Slicer (based on py_client APIs)

Quick Start

Follow the Quick Start guide to build/install AI-Assisted Annotation Client Libraries for C++ and run some basic tools to verify few important functionalities like dextr3D, segmentation, fixPolygon over an existing AI-Assisted Annotation Server.

C++ Client Library provides support for CMake project

For installing py_client:

# 1. build first and install
python setup.py sdist bdist_wheel
python -m pip install dist/aiaa_client-*.whl

# 2. Or just install
python -m pip install -e ./

To use py_client in code:

from py_client import client_api

Contributions

Contributions to NVIDIA AI-Assisted Annotation Client are more than welcome. To contribute make a pull request and follow the guidelines outlined in the CONTRIBUTING document.