RandLA-Net details

About

Short Name RandLA-Net
Long Name RandLA-Net
Website https://github.com/QingyongHu/RandLA-Net
Description We introduce RandLA-Net, an efficient and lightweight neural architecture to directly infer per-point semantics for large-scale point clouds. The key to our approach is to use random point sampling instead of more complex point selection approaches. Although remarkably computation and memory efficient, random sampling can discard key features by chance. To overcome this, we introduce a novel local feature aggregation module to progressively increase the receptive field for each 3D point, thereby effectively preserving geometric details. Extensive experiments show that our RandLA-Net can process 1 million points in a single pass with up to 200X faster than existing approaches. Moreover, our RandLA-Net clearly surpasses state-of-the-art approaches for semantic segmentation on two large-scale benchmarks Semantic3D and SemanticKITTI.
Reference RandLA-Net: Efficient Semantic Segmentation of Large-Scale Point Clouds (CVPR 2020, Oral)
Hardware AMD Ryzen 7 3700X 8-Core Processor × 16, 64G RAM, GeForce RTX 2080 Ti
Used additional training data 0
Last submission 2020-03-19 19:53:50
Is opensource 1
Number of submissions 2

Result

A_IoU OA [s] IoU 1 IoU 2 IoU 3 IoU 4 IoU 5 IoU 6 IoU 7 IoU 8
Result0.7740.9481.000.9560.9140.8660.5150.9570.5150.6980.768
Rank1211728172017861429

Full Confusion Matrix

L1' L2' L3' L4' L5' L6' L7' L8'
L1141263513895607975337829357171919683
L23992651011851426121409322268279252792951
L362134599831628161881286147980
L41324132738061272318490118652012583422594926
L52522837915816185275305865078575085903661
L612574396619221988637261661420234111179703
L7980287528009801278781900042084
L81263002619013145994144065679551