OpenDR toolkit (source distribution)
The aim of OpenDR Project is to develop a modular, open and non-proprietary toolkit for core robotic functionalities by harnessing deep learning to provide advanced perception and cognition capabilities, meeting in this way the general requirements of robotics applications in the applications areas of healthcare, agri-food and agile production. OpenDR provides the means to link the robotics applications to software libraries (deep learning frameworks, e.g., PyTorch and Tensorflow) to the operating environment (ROS). This asset contains the source distribution of the OpenDR toolkit v.3.0.0.
The v3.0.0 of the OpenDR toolkit was released on 4 Dec 2023 and includes the following:
-
New Features:
- Binary High Resolution Learner (#402)
- ROS2 node for EfficientLPS (#404)
- Fall and wave detection ROS nodes (#423)
- Continual SLAM: Adds a new SLAM tool for Continual SLAM (#424)
- Add RGB gesture recognition (#436)
- FSeq2-NMS (#442)
- Intent recognition tool (#443)
- Robotti human detection simulation demo (#451)
- Object Detection 2D Class Filtering (#467)
- RL-based Learner for Active Face Recognition (#473)
- YOLOv5s Inference Demo with Optimized Weights for Agricultural Use (#476)
- Adaptive HR Pose Estimation (#479)
-
Enhancements:
- Wave detection demo based on pose estimation (#394)
- Facial expression recognition demo update (#405)
- Object detection 2d camera demos (#408)
- High Resolution Pose Estimation webcam demo (#409)
- ROS nodes FPS performance measurements (#419)
- Refactoring: pythonic joins in
test_clang_format.py
/test_cppcheck.py
(#455) - Test-tools improvement (#456)
- Adding prompt when transcribe with Whisper (#462)
-
Bug Fixes:
- Fix package creator and sources (#390)
- Lightweight OpenPose tool fixes and improvements (#392)
- Fall Detection - alternative infer input (#397)
- Yolov5 training bugfix (#401)
- Fix the dependency conflict of geffnet installation (#410)
- Fix bug in GEM ROS2 node (#420)
- Fix link to nanodet documentation (#421)
- EfficientLPS panoptic segmentation coloring bug (#426)
- Bump flask from 1.1.2 to 2.3.2 (#430)
- Fix tests on master branch (#438)
- Added unzip installation as base ubuntu dependency and tool tests fixes (#454)
- Active face recognition demo and bug fixes on Face Recognition (#459)
- GPU installation fix (#463)
- Fix ROS1 nodes argparse issue with .launch files (#465)
- Minor fix on yolov5 webcam demo (#466)
- Apply cuDNN init fix to all Object Detectors 2D (#469)
- Updated test_suite_develop.yml based on latest test_suite.yml (#471)
- Fix fmpgmapping (#472)
- Synchronization and bugfixes (#478)
Detailed usage instruction can be found in OpenDR wiki (also included in the AI Asset).