@author: Tibor Camargo
Point cloud processing and analysis of data extracted from multiple Azure Kinect sensors. Includes point cloud registration, filtering, human segmentation and pose estimation.
Currently a personal project, it will be available as a PyPi project around 12/2022 with detailed documentation, installation steps and with a Docker container for reproducing experiments.
There are external dependencies that have not been included yet, such as PyMoCapViewer and OfflineProcessor.
If you are using this software for point cloud processing after MKV extraction:
- Mask R-CNN is used for human segmentation, please download Mask R-CNN weights and configuration here. Do not forget to pass the path to the
.pb
and.pbtxt
accordingly in theextract.py
.
Currently not available