List your company
ResearchFormerly WizWand

Point Cloud Classification on ScanObjectNN OBJ_BG

95.18Overall AccuracyPoint-FEMAE
67.120874.405481.6988.9746Nov 26, 2020Sep 27, 2021Jul 30, 2022Jun 1, 2023Apr 2, 2024Feb 2, 2025Dec 5, 2025
Updated 7mo ago

Evaluation results

MethodLinks
Point-FEMAELearning Paradigm=Single-Modal Self-Supervised Learning, #Params (M)=27.4, Input Data=2k Points2023.1295.18
ReconLearning Paradigm=Cross-Modal Self-Supervised Learning, #Params (M)=44.3, Input Data=2k Points2023.1295.18
Point-MAE w/ LCMPretrain=MPM, #Params(M)=2.7, FLOPs(G)=1.32024.0594.51
I2P-MAELearning Paradigm=Cross-Modal Self-Supervised Learning, #Params (M)=15.3, Input Data=2k Points2023.1294.15
ACT w/ LCMPretrain=MPM, #Params(M)=3.1, FLOPs(G)=2.82024.0594.13
Point-M2AE w/ LCMPretrain=MPM, #Params(M)=2.5, FLOPs(G)=6.72024.0593.83
PointGPT-BPretrain=GPT, #Params(M)=120.5, FLOPs(G)=36.2, Evaluation Setting=Original Paper2024.0593.6
Point-BERT w/ LCMPretrain=MPM, #Params(M)=3.1, FLOPs(G)=2.52024.0593.55
MaskPoint w/ LCMPretrain=MPM, #Params(M)=3.1, FLOPs(G)=2.52024.0593.31
ACTLearning Paradigm=Cross-Modal Self-Supervised Learning, #Params (M)=22.1, Input Data=2k Points2023.1293.29
ACTPretrain=MPM, #Params(M)=22.1, FLOPs(G)=4.8, Evaluation Setting=Original Paper2024.0593.29
PointMambaPretrain=MPM, #Params(M)=12.3, Evaluation Setting=Original Paper2024.0593.29
Point-M2AEPretrain=MPM, #Params(M)=12.9, FLOPs(G)=7.9, Evaluation Setting=Reproduced downstream setting2024.0593.12
LCMPretrain=X, #Params(M)=2.7, FLOPs(G)=1.32024.0592.77
Point-MAEPretrain=MPM, #Params(M)=22.1, FLOPs(G)=4.8, Evaluation Setting=Reproduced downstream setting2024.0592.67
MVTN2020.1192.6
Point-BERTPretrain=MPM, #Params(M)=22.1, FLOPs(G)=4.5, Evaluation Setting=Reproduced downstream setting2024.0592.48
MaskPointPretrain=MPM, #Params(M)=22.1, FLOPs(G)=4.5, Evaluation Setting=Reproduced downstream setting2024.0592.17
ACTPretrain=MPM, #Params(M)=22.1, FLOPs(G)=4.8, Evaluation Setting=Reproduced downstream setting2024.0592.08
TransformerPretrain=X, #Params(M)=22.1, FLOPs(G)=4.8, Evaluation Setting=Reproduced downstream setting2024.0591.95
baselinec=642025.1191.84
MaskFeat3DBackbone=PointViT, Evaluation Protocol=fine-tuning2025.1291.7
Point-M2AELearning Paradigm=Single-Modal Self-Supervised Learning, #Params (M)=15.3, Input Data=2k Points2023.1291.22
Point-M2AE2023.0691.22
Point-MAE w/ IDPTPretrain=MPM, #Params(M)=23.3, FLOPs(G)=7.1, Evaluation Setting=Original Paper2024.0591.22
Point-M2AEPretrain=MPM, #Params(M)=12.9, FLOPs(G)=7.9, Evaluation Setting=Original Paper2024.0591.22
MaskSurfelBackbone=PointViT, Evaluation Protocol=fine-tuning2025.1291.2
Joint-MAELearning Paradigm=Cross-Modal Self-Supervised Learning, Input Data=2k Points2023.1290.94
ExpPoint-MAEunfreeze epoch=2002023.0690.88
Point-MAE w/ DAPTPretrain=MPM, #Params(M)=22.7, FLOPs(G)=5.0, Evaluation Setting=Original Paper2024.0590.88
TAP (Ours)Backbone=Standard Transformer, Pre-training=Generative2023.0790.36
ExpPoint-MAEextra domain adaptation pretraining=true2023.0690.36
Point-MAEBackbone=Standard Transformer, Pre-training=Generative2023.0790.02
Point-MAELearning Paradigm=Single-Modal Self-Supervised Learning, #Params (M)=22.1, Input Data=2k Points2023.1290.02
Point-MAE2023.0690.02
Point-MAEPretrain=MPM, #Params(M)=22.1, FLOPs(G)=4.8, Evaluation Setting=Original Paper2024.0590.02
PointMAEBackbone=PointViT, Evaluation Protocol=fine-tuning2025.1290
MaskDiscrBackbone=PointViT, Evaluation Protocol=fine-tuning2025.1289.7
MaskPointBackbone=Standard Transformer, Pre-training=Generative2023.0789.3
MaskPointLearning Paradigm=Single-Modal Self-Supervised Learning, #Params (M)=22.1, Input Data=2k Points2023.1289.3
MaskPoint2023.0689.3
MaskPointPretrain=MPM, #Params(M)=22.1, FLOPs(G)=4.5, Evaluation Setting=Original Paper2024.0589.3
Inter-MAEPretrain=MPM, #Params(M)=22.1, FLOPs(G)=4.8, Evaluation Setting=Original Paper2024.0588.7
PointMambaPretrain=X, #Params(M)=12.3, Evaluation Setting=Original Paper2024.0588.3
ExpPoint-MAEunfreeze epoch=2502023.0687.44
Point-BERTBackbone=Standard Transformer, Pre-training=Generative2023.0787.43
Point-BERTLearning Paradigm=Single-Modal Self-Supervised Learning, #Params (M)=22.1, Input Data=1k Points2023.1287.43
Point-BERT2023.0687.43
Point-BERTPretrain=MPM, #Params(M)=22.1, FLOPs(G)=4.5, Evaluation Setting=Original Paper2024.0587.43
PointBERTBackbone=PointViT, Evaluation Protocol=fine-tuning2025.1287.4
TransformerPretrain=X, #Params(M)=22.1, FLOPs(G)=4.8, Evaluation Setting=Original Paper2024.0586.75
Foundrys=16, Frozen student=false2025.1186.23
PointCNN2020.1186.1
OcCoBackbone=Standard Transformer, Pre-training=Generative2023.0784.85
DGCNNLearning Paradigm=Supervised Learning Only, #Params (M)=1.8, Input Data=1k Points2023.1282.8
DGCNN2020.1182.8
PointNet++Learning Paradigm=Supervised Learning Only, #Params (M)=1.5, Input Data=1k Points2023.1282.3
PointNet++Pretrain=X, #Params(M)=1.5, FLOPs(G)=1.7, Evaluation Setting=Original Paper2024.0582.3
PointNet ++2020.1182.3
PointViTBackbone=PointViT, Evaluation Protocol=from scratch2025.1279.9
w/o pre-trainingBackbone=Standard Transformer, Pre-training=None2023.0779.86
SpiderCNN2020.1177.1
PointNetLearning Paradigm=Supervised Learning Only, #Params (M)=3.5, Input Data=1k Points2023.1273.3
PointNetPretrain=X, #Params(M)=3.5, FLOPs(G)=0.5, Evaluation Setting=Original Paper2024.0573.3
PointNet2020.1173.3
3DMFV2020.1168.2