Abstract: As 3D LiDAR point cloud and 2D images capture complementary information for autonomous driving, great efforts are made on 3D semantic segmentation using both modalities data. However, they ...

Abstract: Weakly supervised point cloud semantic segmentation methods that require 1% or fewer labels with the aim of realizing almost the same performance as fully supervised approaches have recently ...

2024-11-08: We added 3 new preprocessed example cases for 256x256 resolution inputs. You can now run "dr_strange", "superman", and "minions_stuart" using demo.sh 2024-11-04: LucidFusion now supports ...