ULIP-2 + Point-BERT

Reported on 3 benchmarks across 3 tasks · 1 paper

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

Computer Vision3 results

Shape Representation Of 3D Point CloudsonScanObjectNN
Overall Accuracy· uses extra data· 2023-05-14
89
best: 97.2 (OmniVec2)
ULIP-2: Towards Scalable Multimodal Pre-training for 3D Understanding arXiv:2305.08275
3D Point Cloud ClassificationonScanObjectNN
Overall Accuracy· uses extra data· 2023-05-14
89
best: 97.2 (OmniVec2)
ULIP-2: Towards Scalable Multimodal Pre-training for 3D Understanding arXiv:2305.08275
3D Point Cloud ReconstructiononScanObjectNN
Overall Accuracy· uses extra data· 2023-05-14
89
best: 97.2 (OmniVec2)
ULIP-2: Towards Scalable Multimodal Pre-training for 3D Understanding arXiv:2305.08275