GPr-Net: Geometric Prototypical Network for Point Cloud Few-Shot Learning

Open paper
Benchmark Model Rank Results
few-shot-3d-point-cloud-classification-on-modelnet40-10-way-10-shotGPr-Net + Hyp (512)#27Overall Accuracy: 71.6Standard Deviation: 1.1
few-shot-3d-point-cloud-classification-on-modelnet40-10-way-10-shotGPr-Net + Hyp (1024)#28Overall Accuracy: 70.4Standard Deviation: 1.8
few-shot-3d-point-cloud-classification-on-modelnet40-10-way-10-shotGPr-Net + Euc (512)#29Overall Accuracy: 62.3Standard Deviation: 2.0
few-shot-3d-point-cloud-classification-on-modelnet40-10-way-10-shotGPr-Net + Euc (1024)#30Overall Accuracy: 62.1Standard Deviation: 1.9
few-shot-3d-point-cloud-classification-on-modelnet40-10-way-20-shotGPr-Net + Hyp (512)#26Overall Accuracy: 73.8Standard Deviation: 2.0
few-shot-3d-point-cloud-classification-on-modelnet40-10-way-20-shotGPr-Net + Hyp (1024)#27Overall Accuracy: 72.8Standard Deviation: 1.8
few-shot-3d-point-cloud-classification-on-modelnet40-10-way-20-shotGPr-Net + Euc (1024)#28Overall Accuracy: 63.4Standard Deviation: 2.0
few-shot-3d-point-cloud-classification-on-modelnet40-10-way-20-shotGPr-Net + Euc (512)#29Overall Accuracy: 63.3Standard Deviation: 2.2
few-shot-3d-point-cloud-classification-on-modelnet40-5-way-10-shotGPr-Net + Hyp (512)#25Overall Accuracy: 81.1Standard Deviation: 1.5
few-shot-3d-point-cloud-classification-on-modelnet40-5-way-10-shotGPr-Net + Hyp (1024)#26Overall Accuracy: 80.4Standard Deviation: 0.5
few-shot-3d-point-cloud-classification-on-modelnet40-5-way-10-shotGPr-Net + Euc (1024)#27Overall Accuracy: 74.4Standard Deviation: 2.0
few-shot-3d-point-cloud-classification-on-modelnet40-5-way-10-shotGPr-Net + Euc (512)#28Overall Accuracy: 74.0Standard Deviation: 2.3
few-shot-3d-point-cloud-classification-on-modelnet40-5-way-20-shotGPr-Net + Hyp (512)#25Overall Accuracy: 82.7Standard Deviation: 1.3
few-shot-3d-point-cloud-classification-on-modelnet40-5-way-20-shotGPr-Net + Hyp (1024)#26Overall Accuracy: 82.0Standard Deviation: 0.9
few-shot-3d-point-cloud-classification-on-modelnet40-5-way-20-shotGPr-Net + Euc (1024)#27Overall Accuracy: 75.1Standard Deviation: 2.1
few-shot-3d-point-cloud-classification-on-modelnet40-5-way-20-shotGPr-Net + Euc (512)#28Overall Accuracy: 75.0Standard Deviation: 2.4

Rank counts only results with a code link.