The Open Deep Learning Toolkit for Robotics (OpenDR) 𝘃𝗲𝗿𝘀𝗶𝗼𝗻 𝟭.𝟭 was just released! This new version of the toolkit brings several improvements, including modular installation options, support for newer CUDA versions, and several other improvements! You can download the toolkit through GitHub, pip, and Docker Hub! 

New Features of the toolkit:

• Added end-to-end planning tool (#223).
• Added seq2seq-nms module, along with other custom NMS implementations for 2D object detection.(#232).

Enhancements of the toolkit:

• Added support for modular pip packages allowing tools to be installed separately (#201).
• Simplified the installation process for pip by including the appropriate post-installation scripts (#201).
• Improved the structure of the toolkit by moving io from utils to engine.helper (#201).
• Added support for post-install scripts and opendr dependencies in .ini files (#201).
• Updated toolkit to support CUDA 11.2 and improved GPU support (#215).
• Added a standalone pose-based fall detection tool (#237)

We look forward to receiving your feedback, bug reports, and suggestions for improvements at https://github.com/opendr-eu/opendr !