List your company
ResearchFormerly WizWand

Part Segmentation on ShapeNet In-Context

87.9mIoUPIC-Cat + MICAS
-2.8420.717544.27567.8325Apr 18, 2024May 24, 2024Jun 30, 2024Aug 6, 2024Sep 12, 2024Oct 19, 2024Nov 25, 2024
Updated 7mo ago

Evaluation results

MethodLinks
PIC-Cat + MICASModel training/learning paradigm=In-context learning models, Venues=Ours2024.1187.9
PIC-Sep + MICASModel training/learning paradigm=In-context learning models, Venues=Ours2024.1186.8
PIC++Training Strategy=In-context Learning, Aggregation=-Sep2024.0485.53
PIC++Training Strategy=In-context Learning, Aggregation=-Cat2024.0485.27
PIC-S-CatModel training/learning paradigm=In-context learning models, Venues=Arxiv'242024.1183.8
PIC-S-SepModel training/learning paradigm=In-context learning models, Venues=Arxiv'242024.1183.7
ACTTraining Strategy=Task-Specific2024.0481.24
ACTModel training/learning paradigm=Task-specific models (trained separately), Venues=ICLR'232024.1181.2
PCTModel training/learning paradigm=Task-specific models (trained separately), Venues=CVM'212024.1179.5
PCTTraining Strategy=Task-Specific2024.0479.46
PIC-CatModel training/learning paradigm=In-context learning models, Venues=NeurIPS'232024.1179
PICTraining Strategy=In-context Learning, Aggregation=-Cat2024.0478.95
PointNetModel training/learning paradigm=Task-specific models (trained separately), Venues=CVPR'172024.1177.5
PointNetTraining Strategy=Task-Specific2024.0477.45
DGCNNTraining Strategy=Task-Specific2024.0476.12
DGCNNModel training/learning paradigm=Task-specific models (trained separately), Venues=TOG' 192024.1176.1
PIC-SepModel training/learning paradigm=In-context learning models, Venues=NeurIPS'232024.1175
PICTraining Strategy=In-context Learning, Aggregation=-Sep2024.0474.95
CopyModel training/learning paradigm=In-context learning models2024.1124.2
CopyTraining Strategy=In-context Learning2024.0424.18
I2P-MAEModel training/learning paradigm=Multi-task models: share backbone + multi-task heads, Venues=CVPR'232024.1122.6
PCMamba‡Training Strategy=Multitask‡2024.0419.53
DGCNNModel training/learning paradigm=Multi-task models: share backbone + multi-task heads, Venues=TOG' 192024.1117
DGCNN†Training Strategy=Multitask†2024.0416.95
PCT†Training Strategy=Multitask†2024.0416.71
PCTModel training/learning paradigm=Multi-task models: share backbone + multi-task heads, Venues=CVM'212024.1116.7
PointNet†Training Strategy=Multitask†2024.0415.33
PointNetModel training/learning paradigm=Multi-task models: share backbone + multi-task heads, Venues=CVPR'172024.1115.3
ACTModel training/learning paradigm=Multi-task models: share backbone + multi-task heads, Venues=ICLR'232024.1112.1
ReConModel training/learning paradigm=Multi-task models: share backbone + multi-task heads, Venues=ICML 232024.117.7
PointMAE‡Training Strategy=Multitask‡2024.045.42
Point-MAEModel training/learning paradigm=Multi-task models: share backbone + multi-task heads, Venues=ECCV'222024.115.4
Point-BERTModel training/learning paradigm=In-context learning models, Venues=CVPR 222024.110.7
Point-BERTTraining Strategy=In-context Learning2024.040.65