List your company
ResearchFormerly WizWand

Object Classification on ModelNet40 (Overall Accuracy)

95.1Overall AccuracyADS
83.03686.16889.392.432Apr 22, 2025Jun 30, 2025Sep 7, 2025Nov 16, 2025Jan 24, 2026Apr 3, 2026Jun 12, 2026
Updated 4mo ago

Evaluation results

MethodLinks
ADSPublication=ICCV'23, Points Num.=1k, Evaluation Protocol=Traditional Supervised Learning Only2025.0495.1
Mamba3DBackbone=SSM, Number of trainable parameters (#TP)=16.9 (100%), FLOPs (G)=3.9, Evaluation Protocol=Parameter-Efficient Fine-tuning2026.0594.7
Mamba3D + MantisBackbone=SSM, Number of trainable parameters (#TP)=0.8 (4.7%), FLOPs (G)=4.0, Evaluation Protocol=Parameter-Efficient Fine-tuning2026.0594.7
PointMLPPublication=ICLR'22, Tunable Params.=13.2 M, Points Num.=1k, Evaluation Protocol=Traditional Supervised Learning Only2025.0494.5
RepSurf-UPublication=CVPR'22, Tunable Params.=1.5 M, Points Num.=1k, Evaluation Protocol=Traditional Supervised Learning Only2025.0494.4
Mamba3D + PMABackbone=SSM, Number of trainable parameters (#TP)=1.3 (7.7%), FLOPs (G)=4.7, Evaluation Protocol=Parameter-Efficient Fine-tuning2026.0594.2
ReConBackbone=Attn., Number of trainable parameters (#TP)=43.6 (100%), FLOPs (G)=5.3, Evaluation Protocol=Parameter-Efficient Fine-tuning, Data Modality (Cross-modal vs Single-modal)=cross-modal, Augmentation Strategy=Simple rotational augmentation2026.0594.1
PointNeXtPublication=NeurIPS'22, Tunable Params.=1.4 M, Points Num.=1k, Evaluation Protocol=Traditional Supervised Learning Only2025.0494
Point-M2AEPublication=NeurIPS 22, Tunable Params.=15.3 M, Points Num.=1k, Evaluation Protocol=Self-Supervised Representation Learning (Full Fine-Tuning)2025.0494
Point-M2AEBackbone=Attn., Number of trainable parameters (#TP)=15.3, FLOPs (G)=3.6, Evaluation Protocol=Full Fine-tuning2026.0594
Joint-MAEBackbone=Attn., Evaluation Protocol=Full Fine-tuning2026.0594
ReCon + GAPromptBackbone=Attn., Number of trainable parameters (#TP)=0.6 (1.4%), FLOPs (G)=5.0, Evaluation Protocol=Parameter-Efficient Fine-tuning, Data Modality (Cross-modal vs Single-modal)=cross-modal, Augmentation Strategy=Simple rotational augmentation2026.0594
RECONPublication=ICML 23, Tunable Params.=43.6 M, Points Num.=1k, Evaluation Protocol=Self-Supervised Representation Learning (Full Fine-Tuning)2025.0493.9
MVTNPublication=ICCV'21, Tunable Params.=11.2 M, Points Num.=1k, Evaluation Protocol=Traditional Supervised Learning Only2025.0493.8
MaskPointPublication=ECCV'22, Tunable Params.=22.1 M, Points Num.=1k, Evaluation Protocol=Self-Supervised Representation Learning (Full Fine-Tuning)2025.0493.8
Point-MAEPublication=ECCV'22, Tunable Params.=22.1 M, Points Num.=1k, Evaluation Protocol=Self-Supervised Representation Learning (Full Fine-Tuning)2025.0493.8
Point-MAEBackbone=Attn., Number of trainable parameters (#TP)=22.1 (100%), FLOPs (G)=4.8, Evaluation Protocol=Parameter-Efficient Fine-tuning2026.0593.8
ACTPublication=ICLR'23, Tunable Params.=22.1 M, Points Num.=1k, Evaluation Protocol=Self-Supervised Representation Learning (Full Fine-Tuning)2025.0493.7
ZigzagPointMamba + MantisBackbone=SSM, Number of trainable parameters (#TP)=0.6 (4.9%), FLOPs (G)=4.5, Evaluation Protocol=Parameter-Efficient Fine-tuning2026.0593.7
DC-CCNN++2026.0693.7
ACTBackbone=Attn., Number of trainable parameters (#TP)=22.1, FLOPs (G)=4.8, Evaluation Protocol=Full Fine-tuning, Augmentation Strategy=Simple rotational augmentation2026.0593.6
PointMambaBackbone=SSM, Number of trainable parameters (#TP)=12.3 (100%), FLOPs (G)=3.1, Evaluation Protocol=Parameter-Efficient Fine-tuning, Augmentation Strategy=Simple rotational augmentation2026.0593.6
PointMamba + PMABackbone=SSM, Number of trainable parameters (#TP)=1.1 (8.9%), FLOPs (G)=5.2, Evaluation Protocol=Parameter-Efficient Fine-tuning, Augmentation Strategy=Simple rotational augmentation2026.0593.6
PointMambainput points=1,024, voting mechanism=without2026.0593.6
Point-MAE + DAPTPublication=CVPR 24, Tunable Params.=1.1 M (4.97%), Points Num.=1k, Evaluation Protocol=Self-Supervised Representation Learning (Parameter-Efficient Fine-Tuning)2025.0493.5
ReCon + DAPTBackbone=Attn., Number of trainable parameters (#TP)=1.1 (2.5%), FLOPs (G)=5.0, Evaluation Protocol=Parameter-Efficient Fine-tuning, Data Modality (Cross-modal vs Single-modal)=cross-modal, Augmentation Strategy=Simple rotational augmentation2026.0593.5
PointMamba + MantisBackbone=SSM, Number of trainable parameters (#TP)=0.6 (4.9%), FLOPs (G)=4.0, Evaluation Protocol=Parameter-Efficient Fine-tuning, Augmentation Strategy=Simple rotational augmentation2026.0593.5
DC-CCNN2026.0693.5
I2P-MAEBackbone=Attn., Number of trainable parameters (#TP)=15.3, Evaluation Protocol=Full Fine-tuning, Augmentation Strategy=Simple rotational augmentation2026.0593.4
PCMBackbone=SSM, Number of trainable parameters (#TP)=34.2, Evaluation Protocol=Full Fine-tuning2026.0593.4
Point Mamba (w/ PCT)input points=1,024, voting mechanism=without2026.0593.4
PCMinput points=1,024, voting mechanism=without2026.0593.4
Point-MAE + IDPTPublication=ICCV'23, Tunable Params.=1.7 M (7.69%), Points Num.=1k, Evaluation Protocol=Self-Supervised Representation Learning (Parameter-Efficient Fine-Tuning)2025.0493.3
Point-MAE + Point LoRAPublication=Ours, Tunable Params.=0.77 M (3.43%), Points Num.=1k, Evaluation Protocol=Self-Supervised Representation Learning (Parameter-Efficient Fine-Tuning)2025.0493.3
PointGPT-SBackbone=Attn., Number of trainable parameters (#TP)=29.2, FLOPs (G)=5.7, Evaluation Protocol=Full Fine-tuning, Augmentation Strategy=Simple rotational augmentation2026.0593.3
Point-MAE + IDPTBackbone=Attn., Number of trainable parameters (#TP)=1.7 (7.7%), FLOPs (G)=7.2, Evaluation Protocol=Parameter-Efficient Fine-tuning2026.0593.3
Point-MAE + PointLoRABackbone=Attn., Number of trainable parameters (#TP)=0.8 (3.6%), Evaluation Protocol=Parameter-Efficient Fine-tuning2026.0593.3
Point-BERTPublication=CVPR 22, Tunable Params.=22.1 M, Points Num.=1k, Evaluation Protocol=Self-Supervised Representation Learning (Full Fine-Tuning)2025.0493.2
Point-MAE (Full-FT)Publication=ECCV'22, Tunable Params.=22.1 M (100%), Points Num.=1k, Evaluation Protocol=Self-Supervised Representation Learning (Parameter-Efficient Fine-Tuning)2025.0493.2
Point-MAE + PPT*Publication=arXiv'24, Tunable Params.=1.04 M (4.57%), Points Num.=1k, Evaluation Protocol=Self-Supervised Representation Learning (Parameter-Efficient Fine-Tuning)2025.0493.2
Point-BERTBackbone=Attn., Number of trainable parameters (#TP)=22.1, FLOPs (G)=4.8, Evaluation Protocol=Full Fine-tuning2026.0593.2
ZigzagPointMambaBackbone=SSM, Number of trainable parameters (#TP)=12.3 (100%), FLOPs (G)=3.1, Evaluation Protocol=Parameter-Efficient Fine-tuning2026.0593.2
PCTinput points=1,024, voting mechanism=without2026.0593.2
DensePoint2026.0693.2
PosPool2026.0693.2
PCT2026.0693.2
PointMLP2026.0693.2
DGCNNPublication=TOG' 19, Tunable Params.=1.8 M, Points Num.=1k, Evaluation Protocol=Traditional Supervised Learning Only2025.0492.9
DGCNNBackbone=-, Number of trainable parameters (#TP)=1.8, FLOPs (G)=2.4, Evaluation Protocol=Supervised Learning Only, Data Modality (Cross-modal vs Single-modal)=single-modal2026.0592.9
DGCNNinput points=1,024, voting mechanism=without2026.0592.9
RS-CNNinput points=1,024, voting mechanism=without2026.0592.9
PointASNLinput points=1,024, voting mechanism=without2026.0592.9
RelFlexformer (w/ PCT)input points=1,024, voting mechanism=without2026.0592.9
RS-CNN2026.0692.9
PointASNL2026.0692.9
MLMSPT2026.0692.9
ZigzagPointMamba + PMABackbone=SSM, Number of trainable parameters (#TP)=1.1 (8.9%), FLOPs (G)=5.7, Evaluation Protocol=Parameter-Efficient Fine-tuning2026.0592.8
Point Trans2026.0692.8
SI-MambaBackbone=SSM, Number of trainable parameters (#TP)=12.3, FLOPs (G)=3.6, Evaluation Protocol=Full Fine-tuning2026.0592.7
OctFormerinput points=1,024, voting mechanism=without2026.0592.7
Point Mamba (w/ OctFormer)input points=1,024, voting mechanism=without2026.0592.7
P2Sequenceinput points=1,024, voting mechanism=without2026.0592.6
RelFlexformer (w/ PCT) + PointRoPEinput points=1,024, voting mechanism=without2026.0592.6
PointCNNinput points=1,024, voting mechanism=without2026.0592.5
PointConvinput points=1,024, voting mechanism=without2026.0592.5
PCNNinput points=1,024, voting mechanism=without2026.0592.3
Performer (w/ PCT)input points=1,024, voting mechanism=without2026.0592.3
PointCNNBackbone=-, Number of trainable parameters (#TP)=0.6, FLOPs (G)=0.9, Evaluation Protocol=Supervised Learning Only, Data Modality (Cross-modal vs Single-modal)=single-modal2026.0592.2
OcCoPublication=ICCV'21, Tunable Params.=22.1 M, Points Num.=1k, Evaluation Protocol=Self-Supervised Representation Learning (Full Fine-Tuning)2025.0492.1
PointGridinput points=1,024, voting mechanism=without2026.0592
PointNet++Publication=NeurIPS'17, Tunable Params.=1.5 M, Points Num.=1k, Evaluation Protocol=Traditional Supervised Learning Only2025.0490.7
PointNet++Backbone=-, Number of trainable parameters (#TP)=1.5, FLOPs (G)=1.7, Evaluation Protocol=Supervised Learning Only, Data Modality (Cross-modal vs Single-modal)=single-modal2026.0590.7
PointNet++input points=1,024, voting mechanism=without2026.0590.7
A-SCNinput points=1,024, voting mechanism=without2026.0589.9
PointNetPublication=CVPR'17, Tunable Params.=3.5 M, Points Num.=1k, Evaluation Protocol=Traditional Supervised Learning Only2025.0489.2
PointNetBackbone=-, Number of trainable parameters (#TP)=3.5, FLOPs (G)=0.5, Evaluation Protocol=Supervised Learning Only, Data Modality (Cross-modal vs Single-modal)=cross-modal2026.0589.2
PointNetinput points=1,024, voting mechanism=without2026.0589.2
Spiking PointNet2026.0683.5