Masked Autoencoders for Point Cloud Self-supervised Learning

Benchmark Model Rank Results
point-cloud-segmentation-on-pointcloud-cPointMAE#3mean Corruption Error (mCE): 0.927
few-shot-3d-point-cloud-classification-on-modelnet40-10-way-10-shotPoint-MAE#19Overall Accuracy: 92.6Standard Deviation: 4.1
few-shot-3d-point-cloud-classification-on-modelnet40-10-way-20-shotPoint-MAE#21Overall Accuracy: 95.0Standard Deviation: 3.0
few-shot-3d-point-cloud-classification-on-modelnet40-5-way-10-shotPoint-MAE#21Overall Accuracy: 96.3Standard Deviation: 2.5
few-shot-3d-point-cloud-classification-on-modelnet40-5-way-20-shotPoint-MAE#22Overall Accuracy: 97.8Standard Deviation: 1.8
3d-point-cloud-classification-on-modelnet40Point-MAE#27Overall Accuracy: 94.0
3d-point-cloud-classification-on-scanobjectnnPoint-MAE#53Overall Accuracy: 85.2OBJ-BG (OA): 90.02OBJ-ONLY (OA): 88.29