Pointcept/PointTransformerV3
Point Transformer V3 is a deep learning architecture for processing 3D point clouds, published at CVPR'24 as an Oral paper.

Velocity · 7d
+2.0
★ / day
Trend
→steady
star history
Point Transformer V3 (PTv3) is a neural network architecture designed for 3D point cloud understanding tasks including semantic segmentation and instance segmentation. It extends the Transformer architecture to handle unstructured 3D point data. The repository releases model weights, experiment records, and code integrated into the Pointcept framework.