yuxumin/PoinTr
[ICCV 2021 Oral] PoinTr: Diverse Point Cloud Completion with Geometry-Aware Transformers
This project helps professionals working with 3D scanning or computer vision to reconstruct complete 3D models from partial scans. You input incomplete 3D point cloud data, which could be from a scan where parts were obscured or missed, and it outputs a filled-in, complete 3D model. This is useful for engineers, designers, or researchers who need accurate full models for analysis, manufacturing, or further processing.
820 stars.
Use this if you need to automatically complete missing geometric information in 3D scans or point cloud data to create full, usable 3D models.
Not ideal if you primarily work with 2D images, traditional mesh models, or require manual, precise control over the completion process.
Stars
820
Forks
139
Language
Python
License
MIT
Category
Last pushed
Dec 15, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/computer-vision/yuxumin/PoinTr"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Related tools
drprojects/superpoint_transformer
Official PyTorch implementation of Superpoint Transformer introduced in [ICCV'23] "Efficient 3D...
charlesq34/frustum-pointnets
Frustum PointNets for 3D Object Detection from RGB-D Data
drprojects/DeepViewAgg
[CVPR'22 Best Paper Finalist] Official PyTorch implementation of the method presented in...
facebookresearch/votenet
Deep Hough Voting for 3D Object Detection in Point Clouds
Easonyesheng/A2PM-MESA
[CVPR'24 & TPAMI'26] Area to Point Matching Framework