List your company

Uncertainty-aware Pointmap Prediction on ScanNet++ Indoor

0.0722AURCDeepEns (K=5)
0.067980.0964650.124950.153435May 19, 2026

Evaluation results

MethodLinks
DeepEns (K=5)Inference Type=Sampling, Training Time (h)=100.22026.050.07220.03810.3921
Trust3RInference Type=Single-pass, Training Time (h)=10.52026.050.12330.04440.493
HeteroInference Type=Single-pass, Training Time (h)=9.42026.050.16160.07150.3545
MASt3RInference Type=Single-pass, Training Time (h)=–2026.050.1649—0.2837
MCD (T=16)Inference Type=Sampling, Training Time (h)=–2026.050.17770.09890.1937