PLA

Reported on 4 benchmarks across 1 task · 1 paper · 4 SOTA

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

Computer Vision4 results

3D Open-Vocabulary Instance SegmentationonS3DIS
AP50 Base B6/N6· 2022-11-29
46.9
best: 51.8 (Lowis3D)
SOTA
PLA: Language-Driven Open-Vocabulary 3D Scene Understanding arXiv:2211.16312
3D Open-Vocabulary Instance SegmentationonS3DIS
AP50 Base B8/N4 · 2022-11-29
59
best: 60.8 (Open3DIS)
SOTA
PLA: Language-Driven Open-Vocabulary 3D Scene Understanding arXiv:2211.16312
3D Open-Vocabulary Instance SegmentationonS3DIS
AP50 Novel B6/N6· 2022-11-29
9.8
best: 33 (OpenIns3D)
SOTA
PLA: Language-Driven Open-Vocabulary 3D Scene Understanding arXiv:2211.16312
3D Open-Vocabulary Instance SegmentationonS3DIS
AP50 Novel B8/N4· 2022-11-29
8.6
best: 37 (OpenIns3D)
SOTA
PLA: Language-Driven Open-Vocabulary 3D Scene Understanding arXiv:2211.16312