Tasks SotA Datasets Papers Methods Submit About

Papers With Code 2

A community resource for machine learning research: papers, code, benchmarks, and state-of-the-art results.

Explore

Notable Benchmarks All SotA Datasets Papers Methods

Community

Submit Results About

Data sourced from the PWC Archive (CC-BY-SA 4.0). Built by the community, for the community.

Models/PTv2

PTv2

Reported on 24 benchmarks across 7 tasks · 1 paper · 11 SOTA

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Computer Vision10 results

3D Semantic SegmentationonScanNet++
Top-3 IoU· 2022-10-11
0.688
best: 0.762 (DITR)
SOTA
Point Transformer V2: Grouped Vector Attention and Partition-based Pooling arXiv:2210.05666
LIDAR Semantic SegmentationonnuScenes
test mIoU· 2022-10-11
0.826
best: 0.851 (DITR)
SOTA
Point Transformer V2: Grouped Vector Attention and Partition-based Pooling arXiv:2210.05666
LIDAR Semantic SegmentationonnuScenes
val mIoU· 2022-10-11
0.802
best: 0.842 (DITR)
SOTA
Point Transformer V2: Grouped Vector Attention and Partition-based Pooling arXiv:2210.05666
Shape Representation Of 3D Point CloudsonModelNet40
Mean Accuracy· 2022-10-11
91.6
best: 92.4 (ULIP + PointMLP)
Point Transformer V2: Grouped Vector Attention and Partition-based Pooling arXiv:2210.05666
Shape Representation Of 3D Point CloudsonModelNet40
Overall Accuracy· 2022-10-11
94.2
best: 95.3 (PointGST)
Point Transformer V2: Grouped Vector Attention and Partition-based Pooling arXiv:2210.05666
3D Semantic SegmentationonScanNet++
Top-1 IoU· 2022-10-11
0.445
best: 0.525 (DITR)
Point Transformer V2: Grouped Vector Attention and Partition-based Pooling arXiv:2210.05666
3D Point Cloud ClassificationonModelNet40
Mean Accuracy· 2022-10-11
91.6
best: 92.4 (ULIP + PointMLP)
Point Transformer V2: Grouped Vector Attention and Partition-based Pooling arXiv:2210.05666
3D Point Cloud ClassificationonModelNet40
Overall Accuracy· 2022-10-11
94.2
best: 95.3 (PointGST)
Point Transformer V2: Grouped Vector Attention and Partition-based Pooling arXiv:2210.05666
3D Point Cloud ReconstructiononModelNet40
Mean Accuracy· 2022-10-11
91.6
best: 92.4 (ULIP + PointMLP)
Point Transformer V2: Grouped Vector Attention and Partition-based Pooling arXiv:2210.05666
3D Point Cloud ReconstructiononModelNet40
Overall Accuracy· 2022-10-11
94.2
best: 95.3 (PointGST)
Point Transformer V2: Grouped Vector Attention and Partition-based Pooling arXiv:2210.05666

Medical7 results

Semantic SegmentationonScanNet
val mIoU· 2022-10-11
75.4
best: 80.5 (DITR)
SOTA
Point Transformer V2: Grouped Vector Attention and Partition-based Pooling arXiv:2210.05666
Semantic SegmentationonS3DIS Area5
mIoU· 2022-10-11
72.6
best: 76 (Sonata + PTv3)
SOTA
Point Transformer V2: Grouped Vector Attention and Partition-based Pooling arXiv:2210.05666
Semantic SegmentationonS3DIS Area5
oAcc· 2022-10-11
91.6
best: 93 (Sonata + PTv3)
SOTA
Point Transformer V2: Grouped Vector Attention and Partition-based Pooling arXiv:2210.05666
Semantic SegmentationonScanNet++
Top-3 IoU· 2022-10-11
0.688
best: 0.762 (DITR)
SOTA
Point Transformer V2: Grouped Vector Attention and Partition-based Pooling arXiv:2210.05666
Semantic SegmentationonScanNet
test mIoU· 2022-10-11
75.2
best: 79.8 (PTv3 ARKit LabelMaker)
Point Transformer V2: Grouped Vector Attention and Partition-based Pooling arXiv:2210.05666
Semantic SegmentationonS3DIS Area5
mAcc· 2022-10-11
78
best: 81.6 (Sonata + PTv3)
Point Transformer V2: Grouped Vector Attention and Partition-based Pooling arXiv:2210.05666
Semantic SegmentationonScanNet++
Top-1 IoU· 2022-10-11
0.445
best: 0.525 (DITR)
Point Transformer V2: Grouped Vector Attention and Partition-based Pooling arXiv:2210.05666

Audio7 results

10-shot image generationonScanNet
val mIoU· 2022-10-11
75.4
best: 80.5 (DITR)
SOTA
Point Transformer V2: Grouped Vector Attention and Partition-based Pooling arXiv:2210.05666
10-shot image generationonS3DIS Area5
mIoU· 2022-10-11
72.6
best: 76 (Sonata + PTv3)
SOTA
Point Transformer V2: Grouped Vector Attention and Partition-based Pooling arXiv:2210.05666
10-shot image generationonS3DIS Area5
oAcc· 2022-10-11
91.6
best: 93 (Sonata + PTv3)
SOTA
Point Transformer V2: Grouped Vector Attention and Partition-based Pooling arXiv:2210.05666
10-shot image generationonScanNet++
Top-3 IoU· 2022-10-11
0.688
best: 0.762 (DITR)
SOTA
Point Transformer V2: Grouped Vector Attention and Partition-based Pooling arXiv:2210.05666
10-shot image generationonScanNet
test mIoU· 2022-10-11
75.2
best: 79.8 (PTv3 ARKit LabelMaker)
Point Transformer V2: Grouped Vector Attention and Partition-based Pooling arXiv:2210.05666
10-shot image generationonS3DIS Area5
mAcc· 2022-10-11
78
best: 81.6 (Sonata + PTv3)
Point Transformer V2: Grouped Vector Attention and Partition-based Pooling arXiv:2210.05666
10-shot image generationonScanNet++
Top-1 IoU· 2022-10-11
0.445
best: 0.525 (DITR)
Point Transformer V2: Grouped Vector Attention and Partition-based Pooling arXiv:2210.05666