Towards Compact 3D Representations via Point Feature Enhancement Masked Autoencoders

Benchmark Model Rank Results
few-shot-3d-point-cloud-classification-on-modelnet40-10-way-10-shotPoint-FEMAE#7Overall Accuracy: 94.0
few-shot-3d-point-cloud-classification-on-modelnet40-10-way-20-shotPoint-FEMAE#11Overall Accuracy: 95.8
3d-point-cloud-classification-on-modelnet40Point-FEMAE#12Overall Accuracy: 94.5
3d-point-cloud-classification-on-scanobjectnnPoint-FEMAE#20Overall Accuracy: 90.22OBJ-BG (OA): 95.18OBJ-ONLY (OA): 93.29