Adaptive Point-Prompt Tuning: Fine-Tuning Heterogeneous Foundation Models for 3D Point Cloud Analysis

Benchmark Model Rank Results
3d-part-segmentation-on-shapenet-partAPPT#34Instance Average IoU: 85.9Class Average IoU: 84.0
few-shot-3d-point-cloud-classification-on-modelnet40-10-way-10-shotAPPT (2D)#15Overall Accuracy: 92.7
few-shot-3d-point-cloud-classification-on-modelnet40-10-way-10-shotAPPT (1D (Text))#20Overall Accuracy: 91.5
few-shot-3d-point-cloud-classification-on-modelnet40-10-way-10-shotAPPT (1D (Aud.))#22Overall Accuracy: 91.4
few-shot-3d-point-cloud-classification-on-modelnet40-10-way-20-shotAPPT (2D)#15Overall Accuracy: 95.3
few-shot-3d-point-cloud-classification-on-modelnet40-10-way-20-shotAPPT (1D (Text))#17Overall Accuracy: 95.1
few-shot-3d-point-cloud-classification-on-modelnet40-10-way-20-shotAPPT (1D (Aud.))#20Overall Accuracy: 94.9
few-shot-3d-point-cloud-classification-on-modelnet40-5-way-10-shotAPPT (2D)#13Overall Accuracy: 97.0
few-shot-3d-point-cloud-classification-on-modelnet40-5-way-10-shotAPPT (1D (Text))#17Overall Accuracy: 96.5
few-shot-3d-point-cloud-classification-on-modelnet40-5-way-20-shotAPPT (2D)#4Overall Accuracy: 99.1
few-shot-3d-point-cloud-classification-on-modelnet40-5-way-20-shotAPPT (1D (Text))#6Overall Accuracy: 99.0