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/3D-JEPA

3D-JEPA

Reported on 40 benchmarks across 5 tasks · 1 paper

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

Computer Vision36 results

Shape Representation Of 3D Point CloudsonScanObjectNN
OBJ-BG (OA)· uses extra data· 2024-09-24
93.63
best: 99.48 (PointGST)
3D-JEPA: A Joint Embedding Predictive Architecture for 3D Self-Supervised Representation Learning arXiv:2409.15803
Shape Representation Of 3D Point CloudsonScanObjectNN
OBJ-ONLY (OA)· uses extra data· 2024-09-24
94.49
best: 97.76 (PointGST)
3D-JEPA: A Joint Embedding Predictive Architecture for 3D Self-Supervised Representation Learning arXiv:2409.15803
Shape Representation Of 3D Point CloudsonScanObjectNN
Overall Accuracy· uses extra data· 2024-09-24
89.52
best: 97.2 (OmniVec2)
3D-JEPA: A Joint Embedding Predictive Architecture for 3D Self-Supervised Representation Learning arXiv:2409.15803
Shape Representation Of 3D Point CloudsonModelNet40
Overall Accuracy· 2024-09-24
94
best: 95.3 (PointGST)
3D-JEPA: A Joint Embedding Predictive Architecture for 3D Self-Supervised Representation Learning arXiv:2409.15803
Shape Representation Of 3D Point CloudsonModelNet40 10-way (20-shot)
Overall Accuracy· uses extra data· 2024-09-24
96.3
best: 96.5 (ReCon++)
3D-JEPA: A Joint Embedding Predictive Architecture for 3D Self-Supervised Representation Learning arXiv:2409.15803
Shape Representation Of 3D Point CloudsonModelNet40 10-way (20-shot)
Standard Deviation· uses extra data· 2024-09-24
2.4
best: 13.5 (PointNet)
3D-JEPA: A Joint Embedding Predictive Architecture for 3D Self-Supervised Representation Learning arXiv:2409.15803
Shape Representation Of 3D Point CloudsonModelNet40 5-way (10-shot)
Overall Accuracy· uses extra data· 2024-09-24
97.6
best: 98 (PointGPT)
3D-JEPA: A Joint Embedding Predictive Architecture for 3D Self-Supervised Representation Learning arXiv:2409.15803
Shape Representation Of 3D Point CloudsonModelNet40 5-way (10-shot)
Standard Deviation· uses extra data· 2024-09-24
2
best: 16 (PointNet++)
3D-JEPA: A Joint Embedding Predictive Architecture for 3D Self-Supervised Representation Learning arXiv:2409.15803
Shape Representation Of 3D Point CloudsonModelNet40 10-way (10-shot)
Overall Accuracy· uses extra data· 2024-09-24
94.3
best: 95 (Point-JEPA)
3D-JEPA: A Joint Embedding Predictive Architecture for 3D Self-Supervised Representation Learning arXiv:2409.15803
Shape Representation Of 3D Point CloudsonModelNet40 10-way (10-shot)
Standard Deviation· uses extra data· 2024-09-24
3.6
best: 13.5 (PointNet)
3D-JEPA: A Joint Embedding Predictive Architecture for 3D Self-Supervised Representation Learning arXiv:2409.15803
Shape Representation Of 3D Point CloudsonModelNet40 5-way (20-shot)
Overall Accuracy· uses extra data· 2024-09-24
98.8
best: 99.5 (ReCon++)
3D-JEPA: A Joint Embedding Predictive Architecture for 3D Self-Supervised Representation Learning arXiv:2409.15803
Shape Representation Of 3D Point CloudsonModelNet40 5-way (20-shot)
Standard Deviation· uses extra data· 2024-09-24
0.4
best: 15.5 (PointNet)
3D-JEPA: A Joint Embedding Predictive Architecture for 3D Self-Supervised Representation Learning arXiv:2409.15803
3D Point Cloud ClassificationonScanObjectNN
OBJ-BG (OA)· uses extra data· 2024-09-24
93.63
best: 99.48 (PointGST)
3D-JEPA: A Joint Embedding Predictive Architecture for 3D Self-Supervised Representation Learning arXiv:2409.15803
3D Point Cloud ClassificationonScanObjectNN
OBJ-ONLY (OA)· uses extra data· 2024-09-24
94.49
best: 97.76 (PointGST)
3D-JEPA: A Joint Embedding Predictive Architecture for 3D Self-Supervised Representation Learning arXiv:2409.15803
3D Point Cloud ClassificationonScanObjectNN
Overall Accuracy· uses extra data· 2024-09-24
89.52
best: 97.2 (OmniVec2)
3D-JEPA: A Joint Embedding Predictive Architecture for 3D Self-Supervised Representation Learning arXiv:2409.15803
3D Point Cloud ClassificationonModelNet40
Overall Accuracy· 2024-09-24
94
best: 95.3 (PointGST)
3D-JEPA: A Joint Embedding Predictive Architecture for 3D Self-Supervised Representation Learning arXiv:2409.15803
3D Point Cloud ClassificationonModelNet40 10-way (20-shot)
Overall Accuracy· uses extra data· 2024-09-24
96.3
best: 96.5 (ReCon++)
3D-JEPA: A Joint Embedding Predictive Architecture for 3D Self-Supervised Representation Learning arXiv:2409.15803
3D Point Cloud ClassificationonModelNet40 10-way (20-shot)
Standard Deviation· uses extra data· 2024-09-24
2.4
best: 13.5 (PointNet)
3D-JEPA: A Joint Embedding Predictive Architecture for 3D Self-Supervised Representation Learning arXiv:2409.15803
3D Point Cloud ClassificationonModelNet40 5-way (10-shot)
Overall Accuracy· uses extra data· 2024-09-24
97.6
best: 98 (PointGPT)
3D-JEPA: A Joint Embedding Predictive Architecture for 3D Self-Supervised Representation Learning arXiv:2409.15803
3D Point Cloud ClassificationonModelNet40 5-way (10-shot)
Standard Deviation· uses extra data· 2024-09-24
2
best: 16 (PointNet++)
3D-JEPA: A Joint Embedding Predictive Architecture for 3D Self-Supervised Representation Learning arXiv:2409.15803
3D Point Cloud ClassificationonModelNet40 10-way (10-shot)
Overall Accuracy· uses extra data· 2024-09-24
94.3
best: 95 (Point-JEPA)
3D-JEPA: A Joint Embedding Predictive Architecture for 3D Self-Supervised Representation Learning arXiv:2409.15803
3D Point Cloud ClassificationonModelNet40 10-way (10-shot)
Standard Deviation· uses extra data· 2024-09-24
3.6
best: 13.5 (PointNet)
3D-JEPA: A Joint Embedding Predictive Architecture for 3D Self-Supervised Representation Learning arXiv:2409.15803
3D Point Cloud ClassificationonModelNet40 5-way (20-shot)
Overall Accuracy· uses extra data· 2024-09-24
98.8
best: 99.5 (ReCon++)
3D-JEPA: A Joint Embedding Predictive Architecture for 3D Self-Supervised Representation Learning arXiv:2409.15803
3D Point Cloud ClassificationonModelNet40 5-way (20-shot)
Standard Deviation· uses extra data· 2024-09-24
0.4
best: 15.5 (PointNet)
3D-JEPA: A Joint Embedding Predictive Architecture for 3D Self-Supervised Representation Learning arXiv:2409.15803
3D Point Cloud ReconstructiononScanObjectNN
OBJ-BG (OA)· uses extra data· 2024-09-24
93.63
best: 99.48 (PointGST)
3D-JEPA: A Joint Embedding Predictive Architecture for 3D Self-Supervised Representation Learning arXiv:2409.15803
3D Point Cloud ReconstructiononScanObjectNN
OBJ-ONLY (OA)· uses extra data· 2024-09-24
94.49
best: 97.76 (PointGST)
3D-JEPA: A Joint Embedding Predictive Architecture for 3D Self-Supervised Representation Learning arXiv:2409.15803
3D Point Cloud ReconstructiononScanObjectNN
Overall Accuracy· uses extra data· 2024-09-24
89.52
best: 97.2 (OmniVec2)
3D-JEPA: A Joint Embedding Predictive Architecture for 3D Self-Supervised Representation Learning arXiv:2409.15803
3D Point Cloud ReconstructiononModelNet40
Overall Accuracy· 2024-09-24
94
best: 95.3 (PointGST)
3D-JEPA: A Joint Embedding Predictive Architecture for 3D Self-Supervised Representation Learning arXiv:2409.15803
3D Point Cloud ReconstructiononModelNet40 10-way (20-shot)
Overall Accuracy· uses extra data· 2024-09-24
96.3
best: 96.5 (ReCon++)
3D-JEPA: A Joint Embedding Predictive Architecture for 3D Self-Supervised Representation Learning arXiv:2409.15803
3D Point Cloud ReconstructiononModelNet40 10-way (20-shot)
Standard Deviation· uses extra data· 2024-09-24
2.4
best: 13.5 (PointNet)
3D-JEPA: A Joint Embedding Predictive Architecture for 3D Self-Supervised Representation Learning arXiv:2409.15803
3D Point Cloud ReconstructiononModelNet40 5-way (10-shot)
Overall Accuracy· uses extra data· 2024-09-24
97.6
best: 98 (PointGPT)
3D-JEPA: A Joint Embedding Predictive Architecture for 3D Self-Supervised Representation Learning arXiv:2409.15803
3D Point Cloud ReconstructiononModelNet40 5-way (10-shot)
Standard Deviation· uses extra data· 2024-09-24
2
best: 16 (PointNet++)
3D-JEPA: A Joint Embedding Predictive Architecture for 3D Self-Supervised Representation Learning arXiv:2409.15803
3D Point Cloud ReconstructiononModelNet40 10-way (10-shot)
Overall Accuracy· uses extra data· 2024-09-24
94.3
best: 95 (Point-JEPA)
3D-JEPA: A Joint Embedding Predictive Architecture for 3D Self-Supervised Representation Learning arXiv:2409.15803
3D Point Cloud ReconstructiononModelNet40 10-way (10-shot)
Standard Deviation· uses extra data· 2024-09-24
3.6
best: 13.5 (PointNet)
3D-JEPA: A Joint Embedding Predictive Architecture for 3D Self-Supervised Representation Learning arXiv:2409.15803
3D Point Cloud ReconstructiononModelNet40 5-way (20-shot)
Overall Accuracy· uses extra data· 2024-09-24
98.8
best: 99.5 (ReCon++)
3D-JEPA: A Joint Embedding Predictive Architecture for 3D Self-Supervised Representation Learning arXiv:2409.15803
3D Point Cloud ReconstructiononModelNet40 5-way (20-shot)
Standard Deviation· uses extra data· 2024-09-24
0.4
best: 15.5 (PointNet)
3D-JEPA: A Joint Embedding Predictive Architecture for 3D Self-Supervised Representation Learning arXiv:2409.15803

Medical2 results

Semantic SegmentationonShapeNet-Part
Class Average IoU· uses extra data· 2024-09-24
86.41
best: 87.7 (Feature Geometric Net (FG-Net))
3D-JEPA: A Joint Embedding Predictive Architecture for 3D Self-Supervised Representation Learning arXiv:2409.15803
Semantic SegmentationonShapeNet-Part
Instance Average IoU· uses extra data· 2024-09-24
84.93
best: 89.1 (GeomGCNN)
3D-JEPA: A Joint Embedding Predictive Architecture for 3D Self-Supervised Representation Learning arXiv:2409.15803

Audio2 results

10-shot image generationonShapeNet-Part
Class Average IoU· uses extra data· 2024-09-24
86.41
best: 87.7 (Feature Geometric Net (FG-Net))
3D-JEPA: A Joint Embedding Predictive Architecture for 3D Self-Supervised Representation Learning arXiv:2409.15803
10-shot image generationonShapeNet-Part
Instance Average IoU· uses extra data· 2024-09-24
84.93
best: 89.1 (GeomGCNN)
3D-JEPA: A Joint Embedding Predictive Architecture for 3D Self-Supervised Representation Learning arXiv:2409.15803