Evocargo/Lidar-Annotation-is-All-You-Need
2D road segmentation using lidar data during training
This project helps self-driving car engineers efficiently train their road segmentation models. It takes camera images and sparse 3D lidar data (or existing 2D road segmentation masks) and produces a machine learning model that accurately identifies roads in new camera images. This is ideal for teams working on autonomous vehicle perception systems who need to accurately detect driveable surfaces.
No commits in the last 6 months.
Use this if you need to train an image segmentation model to identify roads using primarily 3D lidar point cloud data, especially when you have limited traditional 2D image annotations.
Not ideal if you are working with non-automotive imaging tasks or if your primary input data is not from lidar sensors combined with cameras.
Stars
43
Forks
3
Language
Python
License
MIT
Category
Last pushed
Dec 21, 2023
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/Evocargo/Lidar-Annotation-is-All-You-Need"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
IGNF/myria3d
Myria3D: Aerial Lidar HD Semantic Segmentation with Deep Learning
PRBonn/lidar-bonnetal
Semantic and Instance Segmentation of LiDAR point clouds for autonomous driving
PRBonn/semantic-kitti-api
SemanticKITTI API for visualizing dataset, processing data, and evaluating results.
venkatasivanaga/FuelDeep3D
R package for LiDAR point-cloud processing and deep-learning inference for 3D fuel mapping.
RWTH-E3D/ifcnet-models
Code for the EG-ICE 2021 Paper "IFCNet: A Benchmark Dataset for IFC Entity Classification"