Label a 3D scene by painting on photos
Fifty-five brush strokes on five photos, and the 3D scene behind them gets its classes: 11.9 % of the points labeled by projection through the cameras, 83.3 % after KD-tree fusion, saved as a labeled PLY and a GLB. CPU only, no training.
The code
Public, no email needed. Read it, fork it, run it, break it.
da_semantic_masking.pyLoad, paint, project, fuse and export, in the article’s 11 cells.interactive_painting.pyThe OpenCV brush tool the script paints with, 1-5 for the class, +/- for the brush.run_headless.pyThe same run with no screen, replaying recorded brush strokes.kids_strokes.jsonThe brush strokes of the reference run, five classes on five photos.
The data and the reference output
Too big to paste out of an article, so tell me where to send them.
reconstruction_data.npzThe KIDS reconstruction from the 3d-models-from-images kit, 9 frames and 1,270,080 points. 44 MB.smart_fused_labels-v2.plyAll 1,270,080 points with their colour and a segment_label field CloudCompare reads as a scalar field. 24 MB.semantic_scene.glbThe labeled scene with the nine camera frustums, for any glTF viewer. 20 MB.
Where should I send them?
One email, and the download links appear right here. You also get them in your inbox so you can come back to this any time.
Here they are
Yours to keep. Run the pipeline on this scene first to reproduce every number in the article, then point it at something of your own.
The same links are in your inbox. If nothing arrives, check spam, then mail me and I will send them by hand.
3D Data Science with Python
Building Accurate Digital Environments with 3D Point Cloud Workflows
This kit is one pipeline. The book is the discipline underneath them: 690 pages from the Python foundations through point cloud processing, meshing and the engineering that keeps a 3D workflow standing up in production. If the script above makes sense and you want the structure they came from, that is where it lives.
See the book or buy it on O’Reilly
Dr. Florent Poux
Founder and Lead Instructor at the 3D Geodata Academy. 15+ years on the automation of reality capture, from point clouds and photogrammetry to spatial AI, and 1,700+ citations across 60+ peer-reviewed publications.
I write on Medium · LinkedIn · YouTube · About · ORCID
These are the materials for Label a 3D scene by painting on photos, one of the tutorials I publish on Medium and on learngeodata.eu. Every kit lives at learngeodata.eu/materials/.
