3D Part Segmentation On Shapenet Part
評価指標
Class Average IoU
Instance Average IoU
評価結果
このベンチマークにおける各モデルのパフォーマンス結果
比較表
| モデル名 | Class Average IoU | Instance Average IoU |
|---|---|---|
| interpolated-convolutional-networks-for-3d | 84.0 | 86.3 |
| point-voxel-transformer-an-efficient-approach | - | 86.5 |
| point-jepa-a-joint-embedding-predictive | 85.8±0.1 | - |
| syncspeccnn-synchronized-spectral-cnn-for-3d | 82.0 | 84.7 |
| pct-point-cloud-transformer | - | 86.4 |
| densepoint-learning-densely-contextual | 84.2 | 86.4 |
| walk-in-the-cloud-learning-curves-for-point | - | 86.8 |
| 3d-jepa-a-joint-embedding-predictive | 86.41 | 84.93 |
| kpconv-flexible-and-deformable-convolution | 85.1 | 86.4 |
| exploiting-local-geometry-for-feature-and | - | 89.1 |
| partnet-a-recursive-part-decomposition | 84.1 | - |
| point2vec-for-self-supervised-representation | 84.6 | 86.3 |
| relation-shape-convolutional-neural-network | - | 86.2 |
| sgpn-similarity-group-proposal-network-for-3d | - | 85.8 |
| point2sequence-learning-the-shape | - | 85.2 |
| pointnet-deep-learning-on-point-sets-for-3d | - | 83.7 |
| モデル 17 | 84.8 | 86.6 |
| pointgrid-a-deep-network-for-3d-shape | 82.2 | 86.4 |
| deltaconv-anisotropic-point-cloud-learning | - | 86.9 |
| generalizing-discrete-convolutions-for | 83.4 | 85.8 |
| pointnet-deep-hierarchical-feature-learning | 81.9 | 85.1 |
| structural-relational-reasoning-of-point | 82.2 | 85.3 |
| anisotropic-separable-set-abstraction-for | - | 86.1 |
| octree-guided-cnn-with-spherical-kernels-for | 83.4 | 86.8 |
| point-cloud-pre-training-by-mixing-and | - | 85.5 |
| point-transformer-1 | 83.7 | 86.6 |
| learning-geometry-disentangled-representation | 85.0 | 86.5 |
| 0-4-dualities | - | 87.0 |
| pointcnn-convolution-on-mathcalx-transformed | 84.6 | 86.14 |
| point-planenet-plane-kernel-based | 82.5 | 85.1 |
| 3d-point-cloud-classification-and | - | 84.3 |
| self-positioning-point-based-transformer-for | 85.4 | 87.2 |
| point-transformer | - | 85.9 |
| so-net-self-organizing-network-for-point | - | 84.9 |
| convolution-in-the-cloud-learning-deformable | 82.1 | 85.1 |
| exploiting-inductive-bias-in-transformer-for | - | 86.2 |
| interpretable-edge-enhancement-and | 85.2 | 87.1 |
| 3d-u-net-learning-dense-volumetric | - | 84.6 |
| point-voxel-cnn-for-efficient-3d-deep | - | 86.2 |
| splatnet-sparse-lattice-networks-for-point | 82.0 | 84.6 |
| take-a-photo-3d-to-2d-generative-pre-training | 85.2 | 86.9 |
| rethinking-masked-representation-learning-for | 85.1 | 86.8 |
| adacrossnet-adaptive-dynamic-loss-weighting | 85.1 | - |
| sagemix-saliency-guided-mixup-for-point | - | 85.7 |
| pointconv-deep-convolutional-networks-on-3d | 82.8 | 85.7 |
| dspoint-dual-scale-point-cloud-recognition | 83.9 | 85.8 |
| odfnet-using-orientation-distribution | 83.3 | 86.5 |
| spidercnn-deep-learning-on-point-sets-with | 82.4 | 85.3 |
| sagemix-saliency-guided-mixup-for-point | - | 85.4 |
| beyond-local-patches-preserving-global-local | - | 88.1 |
| pointnext-revisiting-pointnet-with-improved | 85.2 | 87.1 |
| dynamic-graph-cnn-for-learning-on-point | - | 85.2 |
| fg-net-fast-large-scale-lidar-point | 87.7 | 86.6 |
| escape-from-cells-deep-kd-networks-for-the | 77.4 | 82.3 |
| submanifold-sparse-convolutional-networks | - | 86.0 |
| agcn-adversarial-graph-convolutional-network | 85.7 | 86.9 |
| dense-resolution-network-for-point-cloud | 83.7 | 86.4 |
| spherical-fractal-convolutional-neural | 82.7 | 85.4 |
| point-is-a-vector-a-feature-representation-in | - | 86.9 |
| ckconv-learning-feature-voxelization-for | - | 86.7 |
| spherical-kernel-for-efficient-graph | 84.9 | 86.8 |
| attention-based-point-cloud-edge-sampling | 83.7 | 85.8 |
| beyond-first-impressions-integrating-joint | 82.1 | - |
| attention-based-point-cloud-edge-sampling | 83.1 | 85.6 |
| avs-net-point-sampling-with-adaptive-voxel | 85.7 | 87.3 |
| deltaconv-anisotropic-point-cloud-learning | - | 86.6 |
| p2p-tuning-pre-trained-image-models-for-point | - | 86.5 |