List your company
ResearchFormerly WizWand

3D Object Classification on ScanObjectNN OBJ-ONLY (test)

94.97AccuracyHybridNet
33.121249.178165.23581.2919Nov 29, 2021May 20, 2022Nov 9, 2022May 1, 2023Oct 21, 2023Apr 11, 2024Oct 1, 2024
Updated 8mo ago

Evaluation results

MethodLinks
HybridNetPT=MAP, #P=19.3, #F=4.4, Voting Strategy=w/ vot.2024.1094.97
Mamba3dPT=MAP, #P=16.9, #F=3.9, Voting Strategy=w/ vot.2024.1094.87
Mamba3dPT=Point-MAE, #P=16.9, #F=3.9, Voting Strategy=w/ vot.2024.1094.15
HybridNetPT=MAP, #P=19.3, #F=4.4, Voting Strategy=w/o vot.2024.1093.03
Mamba3dPT=MAP, #P=16.9, #F=3.9, Voting Strategy=w/o vot.2024.1092.75
HybridNetPT=None, #P=19.3, #F=4.4, Voting Strategy=w/ vot.2024.1092.58
Mamba3dPT=None, #P=16.9, #F=3.9, Voting Strategy=w/ vot.2024.1092.43
MVTNPT=None, #P=11.2, #F=43.72024.1092.3
HybridNetPT=None, #P=19.3, #F=4.4, Voting Strategy=w/o vot.2024.1092.28
Mamba3dPT=None, #P=16.9, #F=3.9, Voting Strategy=w/o vot.2024.1092.08
Mamba3dPT=Point-MAE, #P=16.9, #F=3.9, Voting Strategy=w/o vot.2024.1092.08
Mamba3dPT=Point-BERT, #P=16.9, #F=3.9, Voting Strategy=w/o vot.2024.1091.05
TransformerPT=Point-MAE, #P=22.1+1.7, #F=4.82024.1090.02
Point-M2AEPT=Point-M2AE, #P=15.3, #F=3.62024.1088.81
PointMambaPT=Point-MAE, #P=12.3, #F=3.62024.1088.47
TransformerPT=Point-BERT, #P=22.1+1.7, #F=4.82024.1088.3
TransformerPT=Point-MAE, #P=22.1, #F=4.82024.1088.29
Point-BERT2021.1188.12
TransformerPT=MaskPoint, #P=22.1, #F=4.82024.1088.1
PointMambaPT=None, #P=12.3, #F=3.62024.1087.78
DGCNN2021.1186.2
DGCNNPT=None, #P=1.8, #F=2.42024.1086.2
Transformer-OcCo2021.1185.54
TransformerPT=OcCo, #P=22.1, #F=4.82024.1085.54
PointCNN2021.1185.5
PointCNNPT=None, #P=0.62024.1085.5
PointNet++2021.1184.3
PointNet++PT=None, #P=1.5, #F=1.72024.1084.3
Transformer2021.1180.55
TransformerPT=None, #P=22.1, #F=4.82024.1080.55
SpiderCNN2021.1179.5
PointNet2021.1179.2
PointNetPT=None, #P=3.5, #F=0.52024.1079.2
SUGAR (single)Pretrain data=Ensembled (no LVIS), Evaluation protocol=zero-shot, Input Modality=colored point clouds2024.0465.3
SUGAR (multi)Pretrain data=Ensembled (no LVIS), Evaluation protocol=zero-shot, Input Modality=colored point clouds2024.0463.5
SUGAR (multi)Pretrain data=ShapeNet, Evaluation protocol=zero-shot, Input Modality=colored point clouds2024.0455
SUGAR (multi)Pretrain data=ShapeNet, Evaluation protocol=zero-shot, Input Modality=uncolored point clouds2024.0453.5
SUGAR (multi)Pretrain data=Ensembled (no LVIS), Evaluation protocol=zero-shot, Input Modality=uncolored point clouds2024.0453.2
SUGAR (single)Pretrain data=ShapeNet, Evaluation protocol=zero-shot, Input Modality=colored point clouds2024.0452.8
OpenShape-PointBERTPretrain data=ShapeNet, Evaluation protocol=zero-shot, Input Modality=colored point clouds2024.0452
OpenShape-PointBERTPretrain data=Ensembled (no LVIS), Evaluation protocol=zero-shot, Input Modality=colored point clouds2024.0452
OpenShape-PointBERTPretrain data=ShapeNet, Evaluation protocol=zero-shot, Input Modality=uncolored point clouds2024.0451.8
OpenShape-PointBERTPretrain data=Ensembled (no LVIS), Evaluation protocol=zero-shot, Input Modality=uncolored point clouds2024.0450.8
ULIP-PointBERTPretrain data=ShapeNet, Evaluation protocol=zero-shot, Input Modality=uncolored point clouds2024.0449.9
SUGAR (single)Pretrain data=Ensembled (no LVIS), Evaluation protocol=zero-shot, Input Modality=uncolored point clouds2024.0449.6
SUGAR (single)Pretrain data=ShapeNet, Evaluation protocol=zero-shot, Input Modality=uncolored point clouds2024.0448.6
ULIP-PointBERTPretrain data=Ensembled (no LVIS), Evaluation protocol=zero-shot, Input Modality=uncolored point clouds2024.0446
ReConPretrain data=ShapeNet, Evaluation protocol=zero-shot, Input Modality=uncolored point clouds2024.0439.6
CLIP2PointPretrain data=ShapeNet, Evaluation protocol=zero-shot, Input Modality=uncolored point clouds2024.0435.5