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/PointCNN

PointCNN

Reported on 72 benchmarks across 8 tasks · 3 papers · 22 SOTA

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

Computer Vision46 results

Shape Representation Of 3D Point CloudsonScanObjectNN
OBJ-BG (OA)· 2018-01-23
86.1
best: 99.48 (PointGST)
SOTA
PointCNN: Convolution On $\mathcal{X}$-Transformed Points arXiv:1801.07791
Shape Representation Of 3D Point CloudsonScanObjectNN
OBJ-ONLY (OA)· 2018-01-23
85.5
best: 97.76 (PointGST)
SOTA
PointCNN: Convolution On $\mathcal{X}$-Transformed Points arXiv:1801.07791
Shape Representation Of 3D Point CloudsonScanObjectNN
Overall Accuracy· 2018-01-23
78.5
best: 97.2 (OmniVec2)
SOTA
PointCNN: Convolution On $\mathcal{X}$-Transformed Points arXiv:1801.07791
Shape Representation Of 3D Point CloudsonModelNet40 5-way (10-shot)
Overall Accuracy· 2018-01-23
65.41
best: 98 (PointGPT)
SOTA
PointCNN: Convolution On $\mathcal{X}$-Transformed Points arXiv:1801.07791
Instance SegmentationonS3DIS
mAcc· 2018-01-23
75.61
SOTA
PointCNN: Convolution On $\mathcal{X}$-Transformed Points arXiv:1801.07791
3D Point Cloud ClassificationonScanObjectNN
OBJ-BG (OA)· 2018-01-23
86.1
best: 99.48 (PointGST)
SOTA
PointCNN: Convolution On $\mathcal{X}$-Transformed Points arXiv:1801.07791
3D Point Cloud ClassificationonScanObjectNN
OBJ-ONLY (OA)· 2018-01-23
85.5
best: 97.76 (PointGST)
SOTA
PointCNN: Convolution On $\mathcal{X}$-Transformed Points arXiv:1801.07791
3D Point Cloud ClassificationonScanObjectNN
Overall Accuracy· 2018-01-23
78.5
best: 97.2 (OmniVec2)
SOTA
PointCNN: Convolution On $\mathcal{X}$-Transformed Points arXiv:1801.07791
3D Point Cloud ClassificationonModelNet40 5-way (10-shot)
Overall Accuracy· 2018-01-23
65.41
best: 98 (PointGPT)
SOTA
PointCNN: Convolution On $\mathcal{X}$-Transformed Points arXiv:1801.07791
3D Point Cloud ReconstructiononScanObjectNN
OBJ-BG (OA)· 2018-01-23
86.1
best: 99.48 (PointGST)
SOTA
PointCNN: Convolution On $\mathcal{X}$-Transformed Points arXiv:1801.07791
3D Point Cloud ReconstructiononScanObjectNN
OBJ-ONLY (OA)· 2018-01-23
85.5
best: 97.76 (PointGST)
SOTA
PointCNN: Convolution On $\mathcal{X}$-Transformed Points arXiv:1801.07791
3D Point Cloud ReconstructiononScanObjectNN
Overall Accuracy· 2018-01-23
78.5
best: 97.2 (OmniVec2)
SOTA
PointCNN: Convolution On $\mathcal{X}$-Transformed Points arXiv:1801.07791
3D Point Cloud ReconstructiononModelNet40 5-way (10-shot)
Overall Accuracy· 2018-01-23
65.41
best: 98 (PointGPT)
SOTA
PointCNN: Convolution On $\mathcal{X}$-Transformed Points arXiv:1801.07791
3D Instance SegmentationonS3DIS
mAcc· 2018-01-23
75.61
SOTA
PointCNN: Convolution On $\mathcal{X}$-Transformed Points arXiv:1801.07791
Shape Representation Of 3D Point CloudsonScanObjectNN
Mean Accuracy· 2018-01-23
75.1
best: 93.8 (GPSFormer)
PointCNN: Convolution On $\mathcal{X}$-Transformed Points arXiv:1801.07791
Shape Representation Of 3D Point CloudsonModelNet40 5-way (10-shot)
Standard Deviation· 2018-01-23
8.9
best: 16 (PointNet++)
PointCNN: Convolution On $\mathcal{X}$-Transformed Points arXiv:1801.07791
3D Point Cloud ClassificationonScanObjectNN
Mean Accuracy· 2018-01-23
75.1
best: 93.8 (GPSFormer)
PointCNN: Convolution On $\mathcal{X}$-Transformed Points arXiv:1801.07791
3D Point Cloud ClassificationonModelNet40 5-way (10-shot)
Standard Deviation· 2018-01-23
8.9
best: 16 (PointNet++)
PointCNN: Convolution On $\mathcal{X}$-Transformed Points arXiv:1801.07791
3D Point Cloud ReconstructiononScanObjectNN
Mean Accuracy· 2018-01-23
75.1
best: 93.8 (GPSFormer)
PointCNN: Convolution On $\mathcal{X}$-Transformed Points arXiv:1801.07791
3D Point Cloud ReconstructiononModelNet40 5-way (10-shot)
Standard Deviation· 2018-01-23
8.9
best: 16 (PointNet++)
PointCNN: Convolution On $\mathcal{X}$-Transformed Points arXiv:1801.07791
Shape Representation Of 3D Point CloudsonIntrA
F1 score (5-fold)
0.875
best: 0.936 (3DMedPT)
Shape Representation Of 3D Point CloudsonModelNet40
Overall Accuracy
92.2
best: 95.3 (PointGST)
Shape Representation Of 3D Point CloudsonModelNet40 10-way (20-shot)
Overall Accuracy
49.95
best: 96.5 (ReCon++)
Shape Representation Of 3D Point CloudsonModelNet40 10-way (20-shot)
Standard Deviation
7.2
best: 13.5 (PointNet)
Shape Representation Of 3D Point CloudsonModelNet40 10-way (10-shot)
Overall Accuracy
46.6
best: 95 (Point-JEPA)
Shape Representation Of 3D Point CloudsonModelNet40 10-way (10-shot)
Standard Deviation
4.8
best: 13.5 (PointNet)
Shape Representation Of 3D Point CloudsonModelNet40 5-way (20-shot)
Overall Accuracy
68.64
best: 99.5 (ReCon++)
Shape Representation Of 3D Point CloudsonModelNet40 5-way (20-shot)
Standard Deviation
7
best: 15.5 (PointNet)
3D Semantic SegmentationonDALES
Overall Accuracy
97.2
best: 97.8 (KPConv)
3D Semantic SegmentationonDALES
mIoU
58.4
best: 81.1 (KPConv)
3D Point Cloud ClassificationonIntrA
F1 score (5-fold)
0.875
best: 0.936 (3DMedPT)
3D Point Cloud ClassificationonModelNet40
Overall Accuracy
92.2
best: 95.3 (PointGST)
3D Point Cloud ClassificationonModelNet40 10-way (20-shot)
Overall Accuracy
49.95
best: 96.5 (ReCon++)
3D Point Cloud ClassificationonModelNet40 10-way (20-shot)
Standard Deviation
7.2
best: 13.5 (PointNet)
3D Point Cloud ClassificationonModelNet40 10-way (10-shot)
Overall Accuracy
46.6
best: 95 (Point-JEPA)
3D Point Cloud ClassificationonModelNet40 10-way (10-shot)
Standard Deviation
4.8
best: 13.5 (PointNet)
3D Point Cloud ClassificationonModelNet40 5-way (20-shot)
Overall Accuracy
68.64
best: 99.5 (ReCon++)
3D Point Cloud ClassificationonModelNet40 5-way (20-shot)
Standard Deviation
7
best: 15.5 (PointNet)
3D Point Cloud ReconstructiononIntrA
F1 score (5-fold)
0.875
best: 0.936 (3DMedPT)
3D Point Cloud ReconstructiononModelNet40
Overall Accuracy
92.2
best: 95.3 (PointGST)
3D Point Cloud ReconstructiononModelNet40 10-way (20-shot)
Overall Accuracy
49.95
best: 96.5 (ReCon++)
3D Point Cloud ReconstructiononModelNet40 10-way (20-shot)
Standard Deviation
7.2
best: 13.5 (PointNet)
3D Point Cloud ReconstructiononModelNet40 10-way (10-shot)
Overall Accuracy
46.6
best: 95 (Point-JEPA)
3D Point Cloud ReconstructiononModelNet40 10-way (10-shot)
Standard Deviation
4.8
best: 13.5 (PointNet)
3D Point Cloud ReconstructiononModelNet40 5-way (20-shot)
Overall Accuracy
68.64
best: 99.5 (ReCon++)
3D Point Cloud ReconstructiononModelNet40 5-way (20-shot)
Standard Deviation
7
best: 15.5 (PointNet)

Medical14 results

Semantic SegmentationonIntrA
DSC (V)· 2018-01-23
96.62
best: 97.29 (3DMedPT)
SOTA
PointCNN: Convolution On $\mathcal{X}$-Transformed Points arXiv:1801.07791
Semantic SegmentationonIntrA
IoU (V)· 2018-01-23
93.59
best: 94.82 (3DMedPT)
SOTA
PointCNN: Convolution On $\mathcal{X}$-Transformed Points arXiv:1801.07791
Semantic SegmentationonShapeNet-Part
Class Average IoU· 2018-01-23
84.6
best: 87.7 (Feature Geometric Net (FG-Net))
SOTA
PointCNN: Convolution On $\mathcal{X}$-Transformed Points arXiv:1801.07791
Semantic SegmentationonShapeNet-Part
Instance Average IoU· 2018-01-23
86.14
best: 89.1 (GeomGCNN)
SOTA
PointCNN: Convolution On $\mathcal{X}$-Transformed Points arXiv:1801.07791
Semantic SegmentationonS3DIS Area5
mIoU· 2020-12-16
57.3
best: 76 (Sonata + PTv3)
Point Transformer arXiv:2012.09164
Semantic SegmentationonS3DIS
Mean IoU· 2020-12-16
65.4
best: 82.3 (Sonata + PTv3)
Point Transformer arXiv:2012.09164
Semantic SegmentationonS3DIS
Mean IoU· 2020-12-16
65.4
best: 82.3 (Sonata + PTv3)
Point Transformer arXiv:2012.09164
Semantic SegmentationonS3DIS
oAcc· 2019-04-16
88.1
best: 93.3 (Sonata + PTv3)
A-CNN: Annularly Convolutional Neural Networks on Point Clouds arXiv:1904.08017
Semantic SegmentationonIntrA
DSC (A)· 2018-01-23
81.74
best: 89.71 (3DMedPT)
PointCNN: Convolution On $\mathcal{X}$-Transformed Points arXiv:1801.07791
Semantic SegmentationonIntrA
IoU (A)· 2018-01-23
74.11
best: 82.39 (3DMedPT)
PointCNN: Convolution On $\mathcal{X}$-Transformed Points arXiv:1801.07791
Semantic SegmentationonScanNet
test mIoU
45.8
best: 79.8 (PTv3 ARKit LabelMaker)
Semantic SegmentationonS3DIS Area5
oAcc
85.9
best: 93 (Sonata + PTv3)
Semantic SegmentationonDALES
Overall Accuracy
97.2
best: 97.8 (KPConv)
Semantic SegmentationonDALES
mIoU
58.4
best: 81.1 (KPConv)

Audio14 results

10-shot image generationonIntrA
DSC (V)· 2018-01-23
96.62
best: 97.29 (3DMedPT)
SOTA
PointCNN: Convolution On $\mathcal{X}$-Transformed Points arXiv:1801.07791
10-shot image generationonIntrA
IoU (V)· 2018-01-23
93.59
best: 94.82 (3DMedPT)
SOTA
PointCNN: Convolution On $\mathcal{X}$-Transformed Points arXiv:1801.07791
10-shot image generationonShapeNet-Part
Class Average IoU· 2018-01-23
84.6
best: 87.7 (Feature Geometric Net (FG-Net))
SOTA
PointCNN: Convolution On $\mathcal{X}$-Transformed Points arXiv:1801.07791
10-shot image generationonShapeNet-Part
Instance Average IoU· 2018-01-23
86.14
best: 89.1 (GeomGCNN)
SOTA
PointCNN: Convolution On $\mathcal{X}$-Transformed Points arXiv:1801.07791
10-shot image generationonS3DIS Area5
mIoU· 2020-12-16
57.3
best: 76 (Sonata + PTv3)
Point Transformer arXiv:2012.09164
10-shot image generationonS3DIS
Mean IoU· 2020-12-16
65.4
best: 82.3 (Sonata + PTv3)
Point Transformer arXiv:2012.09164
10-shot image generationonS3DIS
Mean IoU· 2020-12-16
65.4
best: 82.3 (Sonata + PTv3)
Point Transformer arXiv:2012.09164
10-shot image generationonS3DIS
oAcc· 2019-04-16
88.1
best: 93.3 (Sonata + PTv3)
A-CNN: Annularly Convolutional Neural Networks on Point Clouds arXiv:1904.08017
10-shot image generationonIntrA
DSC (A)· 2018-01-23
81.74
best: 89.71 (3DMedPT)
PointCNN: Convolution On $\mathcal{X}$-Transformed Points arXiv:1801.07791
10-shot image generationonIntrA
IoU (A)· 2018-01-23
74.11
best: 82.39 (3DMedPT)
PointCNN: Convolution On $\mathcal{X}$-Transformed Points arXiv:1801.07791
10-shot image generationonScanNet
test mIoU
45.8
best: 79.8 (PTv3 ARKit LabelMaker)
10-shot image generationonS3DIS Area5
oAcc
85.9
best: 93 (Sonata + PTv3)
10-shot image generationonDALES
Overall Accuracy
97.2
best: 97.8 (KPConv)
10-shot image generationonDALES
mIoU
58.4
best: 81.1 (KPConv)