Semantic Segmentation On Scannet
Metriken
test mIoU
val mIoU
Ergebnisse
Leistungsergebnisse verschiedener Modelle zu diesem Benchmark
Vergleichstabelle
| Modellname | test mIoU | val mIoU |
|---|---|---|
| lsk3dnet-towards-effective-and-efficient-3d | 75.5 | 75.7 |
| masked-scene-contrast-a-scalable-framework | - | 75.5 |
| fully-convolutional-point-networks-for-large | 44.7 | - |
| tangent-convolutions-for-dense-prediction-in | 44.2 | - |
| bfanet-revisiting-3d-semantic-segmentation | - | 78.0 |
| point-transformer-v2-grouped-vector-attention | 75.2 | 75.4 |
| pointcnn-convolution-on-x-transformed-points | 45.8 | - |
| serialized-point-mamba-a-serialized-point | - | 76.8 |
| texturenet-consistent-local-parametrizations | 56.6 | - |
| fg-net-fast-large-scale-lidar-point | 69.0 | - |
| octformer-octree-based-transformers-for-3d | 76.6 | 75.7 |
| 3dmv-joint-3d-multi-view-prediction-for-3d | 48.4 | - |
| pointnet-deep-hierarchical-feature-learning | 33.9 | 53.5 |
| kpconvx-modernizing-kernel-point-convolution | - | 76.3 |
| bidirectional-projection-network-for-cross | 74.9 | 73.9 |
| stratified-transformer-for-3d-point-cloud | 73.7 | 74.3 |
| towards-large-scale-3d-representation | 76.6 | 76.4 |
| oneformer3d-one-transformer-for-unified-point | - | 76.6 |
| arkit-labelmaker-a-new-scale-for-indoor-3d | 79.8 | 79.1 |
| avs-net-point-sampling-with-adaptive-voxel | - | 76.1 |
| oa-cnns-omni-adaptive-sparse-cnns-for-3d | 75.6 | 76.1 |
| kpconv-flexible-and-deformable-convolution | 68.0 | 69.2 |
| sonata-self-supervised-learning-of-reliable | - | 79.4 |
| swin3d-a-pretrained-transformer-backbone-for | 77.9 | 77.5 |
| decoupled-local-aggregation-for-point-cloud | - | 75.9 |
| panopticfusion-online-volumetric-semantic | 52.9 | - |
| 3d-semantic-segmentation-with-submanifold | 72.5 | 69.3 |
| learning-inner-group-relations-on-point | 68.2 | - |
| ponderv2-pave-the-way-for-3d-foundataion | 78.5 | 77.0 |
| splatnet-sparse-lattice-networks-for-point | 39.3 | - |
| similarity-aware-fusion-network-for-3d | 65.4 | - |
| odin-a-single-model-for-2d-and-3d-perception | 74.4 | 77.8 |
| mix3d-out-of-context-data-augmentation-for-3d | 78.1 | 73.6 |
| pointconv-deep-convolutional-networks-on-3d | 55.6 | 61.0 |
| scannet-richly-annotated-3d-reconstructions | 30.6 | - |
| dino-in-the-room-leveraging-2d-foundation | 79.7 | 80.5 |
| pointhr-exploring-high-resolution | 76.6 | 75.4 |
| a-unified-query-based-paradigm-for-point | 74.3 | 75.3 |
| convolutional-neural-networks-on-3d-surfaces | 44.2 | - |
| virtual-multi-view-fusion-for-3d-semantic | 74.6 | - |
| mamba24-8d-enhancing-global-interaction-in | - | 77.6 |
| point-transformer-v3-simpler-faster-stronger | 79.4 | 78.6 |
| panopticndt-efficient-and-robust-panoptic | 68.1 | 68.39 |
| 4d-spatio-temporal-convnets-minkowski | 73.4 | 72.2 |
| o-cnn-octree-based-convolutional-neural | 76.2 | 74.0 |