Asymmetric Dual Self-Distillation for 3D Self-Supervised Representation Learning

Open paper Other version
Benchmark Model Rank Results
few-shot-3d-point-cloud-classification-on-modelnet40-10-way-10-shotAsymDSD-S#3Overall Accuracy: 94.4
3d-point-cloud-classification-on-scanobjectnnAsymDSD-B* (no voting)#4Overall Accuracy: 93.72OBJ-BG (OA): 96.73OBJ-ONLY (OA): 94.32
3d-point-cloud-classification-on-scanobjectnnAsymDSD-S (no voting)#13Overall Accuracy: 90.53OBJ-BG (OA): 94.32OBJ-ONLY (OA): 91.91
3d-point-cloud-classification-on-modelnet40AsymDSD-B* (no voting)#9Overall Accuracy: 94.7
few-shot-3d-point-cloud-classification-on-modelnet40-10-way-20-shotAsymDSD-S#12Overall Accuracy: 95.8
few-shot-3d-point-cloud-classification-on-modelnet40-5-way-20-shotAsymDSD-S#13Overall Accuracy: 98.8
few-shot-3d-point-cloud-classification-on-modelnet40-5-way-10-shotAsymDSD-S#22Overall Accuracy: 96.1

Rank counts only results with a code link.