zhulf0804/PREDATOR
An inofficial PyTorch implementation of PREDATOR based on KPConv.
This project helps 3D scanning and computer vision engineers precisely align multiple 3D point cloud scans, especially when the scans have very little common overlap. It takes two or more raw 3D point cloud datasets as input and outputs a single, accurately registered (aligned) 3D model. This is for professionals working with 3D data from environments or objects scanned from different perspectives.
No commits in the last 6 months.
Use this if you need to combine multiple 3D scans into one coherent model, and those scans were taken from very different angles, resulting in minimal shared areas between them.
Not ideal if your 3D scans already have significant overlap, as simpler and faster registration methods might suffice.
Stars
17
Forks
1
Language
C++
License
—
Category
Last pushed
Oct 17, 2021
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/computer-vision/zhulf0804/PREDATOR"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
drprojects/superpoint_transformer
Official PyTorch implementation of Superpoint Transformer introduced in [ICCV'23] "Efficient 3D...
yuxumin/PoinTr
[ICCV 2021 Oral] PoinTr: Diverse Point Cloud Completion with Geometry-Aware Transformers
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