3D Instance Segmentation On Scannetv2
Métriques
mAP
mAP @ 50
mRec
Résultats
Résultats de performance de divers modèles sur ce benchmark
Tableau comparatif
| Nom du modèle | mAP | mAP @ 50 | mRec |
|---|---|---|---|
| learning-object-bounding-boxes-for-3d | 25.3 | 48.8 | 47.6 |
| masc-multi-scale-affinity-with-sparse | - | 45.9 | - |
| msta3d-multi-scale-twin-attention-for-3d-1 | 56.9 | 79.5 | 74.1 |
| learning-regional-purity-for-instance | 42.8 | 64.2 | - |
| 3d-u-net-learning-dense-volumetric | - | 31.9 | - |
| serialized-point-mamba-a-serialized-point | 40.0 | - | - |
| hida-towards-holistic-indoor-understanding | 43.6 | 63.5 | - |
| instance-segmentation-in-3d-scenes-using | 50.6 | 69.8 | - |
| spherical-mask-coarse-to-fine-3d-point-cloud | 61.6 | 81.2 | - |
| pointgroup-dual-set-point-grouping-for-3d | 40.7 | 63.6 | - |
| 3d-mpa-multi-proposal-aggregation-for-3d | 35.3 | 59.1 | 61.1 |
| mask-attention-free-transformer-for-3d | 59.6 | - | - |
| hierarchical-aggregation-for-3d-instance | 45.7 | 69.9 | - |
| odin-a-single-model-for-2d-and-3d-perception | 50.0 | 71.0 | - |
| oneformer3d-one-transformer-for-unified-point | 56.6 | 80.1 | - |
| isbnet-a-3d-point-cloud-instance-segmentation | 55.9 | 76.3 | - |
| beyond-the-final-layer-hierarchical-query | 60.6 | 81.0 | - |
| 3d-instances-as-1d-kernels | 53.2 | 71.8 | - |
| top-down-beats-bottom-up-in-3d-instance | 48.9 | 75.1 | - |
| learning-gaussian-instance-segmentation-in | - | 63.8 | - |
| 3d-sis-3d-semantic-instance-segmentation-of | - | 38.2 | - |
| 4dcontrast-contrastive-learning-with-dynamic | - | 57.6 | - |
| divide-and-conquer-3d-point-cloud-instance | 57.3 | 74.7 | - |
| superpoint-transformer-for-3d-scene-instance | 54.9 | 77.0 | - |
| softgroup-for-3d-instance-segmentation-on | 50.4 | 76.1 | - |
| mask3d-for-3d-semantic-instance-segmentation | 55.2 | 78.0 | - |
| rwseg-cross-graph-competing-random-walks-for | 34.8 | 56.7 | - |
| learning-and-memorizing-representative | - | 31 | - |
| maskgroup-hierarchical-point-grouping-and | 43.4 | 66.4 | - |
| query-refinement-transformer-for-3d-instance | 58.3 | 78.7 | - |
| masc-multi-scale-affinity-with-sparse | 25.4 | 44.7 | - |