Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

LERF

Benchmarks

Task NameDataset NameSOTA ResultTrend
Open-Vocabulary 3D Scene SegmentationLeRF-mask
Figurines mIoU90.8
17
3D Semantic SegmentationLERF (test)
mIoU62.1
13
3D Scene ReconstructionLERF average across four scenes
PSNR24.02
12
3D Open-vocabulary SegmentationLERF-style Dataset bed scene (test)
mIoU89.5
8
Open-vocabulary 3D Scene UnderstandingLERF
Feature Distillation Time (h)1
7
Open-vocabulary semantic segmentationLeRF-OVS
mIoU45.8
6
3D object localizationLERF (test)
RAMEN Score0.732
3
3D Scene ReconstructionLERF
Training Time (min)29.5
2
Showing 8 of 8 rows