This tool can import 3d point-clouds / camera poses / pictures and tag them:
- imports from MVE scenes
- renders point-cloud and cameras in 3d using opengl
- generates/refines point-cloud selections by hand (spherical selections)
- reprojects/exports point-cloud and selections into images (masks)
This code was developed in the scope of a research project:
Deep-Learning Image Segmentation - Towards Tea Leaves Harvesting by Automomous Machine
- Python 3.x
- QT Pyside2 >= 5.11
- OpenGL >= 2
- Precomputed MVE scenes
python3 app.py
GPLv3