sairajk/asia
[SIGGRAPH Asia 2025] "ASIA: Adaptive 3D Segmentation using Few Image Annotations ".
This project helps 3D artists, game developers, and CAD designers efficiently segment complex 3D models into distinct parts. By providing a few simple 2D image annotations, you can define specific regions, even non-standard or hard-to-describe ones, and get a precisely segmented 3D model as output. This is ideal for anyone working with 3D assets who needs to break them down for texturing, rigging, or further editing.
Use this if you need to precisely segment 3D models into specific parts using minimal annotation effort on 2D images, rather than complex 3D marking or ambiguous text descriptions.
Not ideal if you primarily work with existing text-describable semantic segments, or if you prefer manual 3D annotation over image-based methods.
Stars
25
Forks
—
Language
—
License
—
Category
Last pushed
Feb 14, 2026
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/sairajk/asia"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
daavoo/pyntcloud
pyntcloud is a Python library for working with 3D point clouds.
yangyanli/PointCNN
PointCNN: Convolution On X-Transformed Points (NeurIPS 2018)
torch-points3d/torch-points3d
Pytorch framework for doing deep learning on point clouds.
yogeshhk/MidcurveNN
Computation of Midcurve of Thin Polygons using Neural Networks
charlesq34/pointnet2
PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Space