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/PointCLIP V2

PointCLIP V2

Reported on 18 benchmarks across 5 tasks · 1 paper · 18 SOTA

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

Computer Vision18 results

Shape Representation Of 3D Point CloudsonScanObjectNN
OBJ_BG Accuracy(%)· uses extra data· 2022-11-21
41.22
SOTA
PointCLIP V2: Prompting CLIP and GPT for Powerful 3D Open-world Learning arXiv:2211.11682
Shape Representation Of 3D Point CloudsonScanObjectNN
OBJ_ONLY Accuracy(%)· uses extra data· 2022-11-21
50.09
best: 65.4 (ReCon++)
SOTA
PointCLIP V2: Prompting CLIP and GPT for Powerful 3D Open-world Learning arXiv:2211.11682
Shape Representation Of 3D Point CloudsonScanObjectNN
PB_T50_RS Accuracy (%)· uses extra data· 2022-11-21
35.36
SOTA
PointCLIP V2: Prompting CLIP and GPT for Powerful 3D Open-world Learning arXiv:2211.11682
Shape Representation Of 3D Point CloudsonModelNet40
Accuracy (%)· uses extra data· 2022-11-21
64.22
best: 88.2 (Uni3D)
SOTA
PointCLIP V2: Prompting CLIP and GPT for Powerful 3D Open-world Learning arXiv:2211.11682
Shape Representation Of 3D Point CloudsonModelNet10
Accuracy (%)· uses extra data· 2022-11-21
73.13
best: 75.6 (ReCon)
SOTA
PointCLIP V2: Prompting CLIP and GPT for Powerful 3D Open-world Learning arXiv:2211.11682
3D Point Cloud ClassificationonScanObjectNN
OBJ_BG Accuracy(%)· uses extra data· 2022-11-21
41.22
SOTA
PointCLIP V2: Prompting CLIP and GPT for Powerful 3D Open-world Learning arXiv:2211.11682
3D Point Cloud ClassificationonScanObjectNN
OBJ_ONLY Accuracy(%)· uses extra data· 2022-11-21
50.09
best: 65.4 (ReCon++)
SOTA
PointCLIP V2: Prompting CLIP and GPT for Powerful 3D Open-world Learning arXiv:2211.11682
3D Point Cloud ClassificationonScanObjectNN
PB_T50_RS Accuracy (%)· uses extra data· 2022-11-21
35.36
SOTA
PointCLIP V2: Prompting CLIP and GPT for Powerful 3D Open-world Learning arXiv:2211.11682
3D Point Cloud ClassificationonModelNet40
Accuracy (%)· uses extra data· 2022-11-21
64.22
best: 88.2 (Uni3D)
SOTA
PointCLIP V2: Prompting CLIP and GPT for Powerful 3D Open-world Learning arXiv:2211.11682
3D Point Cloud ClassificationonModelNet10
Accuracy (%)· uses extra data· 2022-11-21
73.13
best: 75.6 (ReCon)
SOTA
PointCLIP V2: Prompting CLIP and GPT for Powerful 3D Open-world Learning arXiv:2211.11682
Training-free 3D Point Cloud ClassificationonModelNet40
Accuracy (%)· uses extra data· 2022-11-21
64.2
best: 85.3 (Point-GN)
SOTA
PointCLIP V2: Prompting CLIP and GPT for Powerful 3D Open-world Learning arXiv:2211.11682
Training-free 3D Point Cloud ClassificationonScanObjectNN
Accuracy (%)· uses extra data· 2022-11-21
35.4
best: 86.4 (Point-GN)
SOTA
PointCLIP V2: Prompting CLIP and GPT for Powerful 3D Open-world Learning arXiv:2211.11682
Training-free 3D Part SegmentationonShapeNet-Part
mIoU· uses extra data· 2022-11-21
48.4
best: 74 (Point-NN)
SOTA
PointCLIP V2: Prompting CLIP and GPT for Powerful 3D Open-world Learning arXiv:2211.11682
3D Point Cloud ReconstructiononScanObjectNN
OBJ_BG Accuracy(%)· uses extra data· 2022-11-21
41.22
SOTA
PointCLIP V2: Prompting CLIP and GPT for Powerful 3D Open-world Learning arXiv:2211.11682
3D Point Cloud ReconstructiononScanObjectNN
OBJ_ONLY Accuracy(%)· uses extra data· 2022-11-21
50.09
best: 65.4 (ReCon++)
SOTA
PointCLIP V2: Prompting CLIP and GPT for Powerful 3D Open-world Learning arXiv:2211.11682
3D Point Cloud ReconstructiononScanObjectNN
PB_T50_RS Accuracy (%)· uses extra data· 2022-11-21
35.36
SOTA
PointCLIP V2: Prompting CLIP and GPT for Powerful 3D Open-world Learning arXiv:2211.11682
3D Point Cloud ReconstructiononModelNet40
Accuracy (%)· uses extra data· 2022-11-21
64.22
best: 88.2 (Uni3D)
SOTA
PointCLIP V2: Prompting CLIP and GPT for Powerful 3D Open-world Learning arXiv:2211.11682
3D Point Cloud ReconstructiononModelNet10
Accuracy (%)· uses extra data· 2022-11-21
73.13
best: 75.6 (ReCon)
SOTA
PointCLIP V2: Prompting CLIP and GPT for Powerful 3D Open-world Learning arXiv:2211.11682