3D Medical Point Transformer: Introducing Convolution to Attention Networks for Medical Point Cloud Analysis

Benchmark Model Rank Results
3d-part-segmentation-on-intra3DMedPT#1IoU (V): 94.82IoU (A): 82.39DSC (V): 97.29DSC (A): 89.71
3d-point-cloud-classification-on-intra3DMedPT#1F1 score (5-fold): 0.936
3d-point-cloud-classification-on-modelnet403DMedPT#62Overall Accuracy: 93.4Number of params: 1.54M