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/PonderV2 + SparseUNet

PonderV2 + SparseUNet

Reported on 22 benchmarks across 3 tasks · 1 paper · 9 SOTA

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

Medical10 results

Semantic SegmentationonScanNet
test mIoU· uses extra data· 2023-10-12
78.5
best: 79.8 (PTv3 ARKit LabelMaker)
SOTA
PonderV2: Pave the Way for 3D Foundation Model with A Universal Pre-training Paradigm arXiv:2310.08586
Semantic SegmentationonS3DIS
Mean IoU· uses extra data· 2023-10-12
79.9
best: 82.3 (Sonata + PTv3)
SOTA
PonderV2: Pave the Way for 3D Foundation Model with A Universal Pre-training Paradigm arXiv:2310.08586
Semantic SegmentationonS3DIS
oAcc· uses extra data· 2023-10-12
92.5
best: 93.3 (Sonata + PTv3)
SOTA
PonderV2: Pave the Way for 3D Foundation Model with A Universal Pre-training Paradigm arXiv:2310.08586
Semantic SegmentationonScanNet200
test mIoU· uses extra data· 2023-10-12
34.6
best: 44.9 (DITR)
SOTA
PonderV2: Pave the Way for 3D Foundation Model with A Universal Pre-training Paradigm arXiv:2310.08586
Semantic SegmentationonScanNet
val mIoU· uses extra data· 2023-10-12
77
best: 80.5 (DITR)
PonderV2: Pave the Way for 3D Foundation Model with A Universal Pre-training Paradigm arXiv:2310.08586
Semantic SegmentationonS3DIS Area5
mAcc· uses extra data· 2023-10-12
79
best: 81.6 (Sonata + PTv3)
PonderV2: Pave the Way for 3D Foundation Model with A Universal Pre-training Paradigm arXiv:2310.08586
Semantic SegmentationonS3DIS Area5
mIoU· uses extra data· 2023-10-12
73.2
best: 76 (Sonata + PTv3)
PonderV2: Pave the Way for 3D Foundation Model with A Universal Pre-training Paradigm arXiv:2310.08586
Semantic SegmentationonS3DIS Area5
oAcc· uses extra data· 2023-10-12
92.2
best: 93 (Sonata + PTv3)
PonderV2: Pave the Way for 3D Foundation Model with A Universal Pre-training Paradigm arXiv:2310.08586
Semantic SegmentationonS3DIS
mAcc· uses extra data· 2023-10-12
86.5
best: 89.9 (Sonata + PTv3)
PonderV2: Pave the Way for 3D Foundation Model with A Universal Pre-training Paradigm arXiv:2310.08586
Semantic SegmentationonScanNet200
val mIoU· uses extra data· 2023-10-12
32.3
best: 41.2 (DITR)
PonderV2: Pave the Way for 3D Foundation Model with A Universal Pre-training Paradigm arXiv:2310.08586

Audio10 results

10-shot image generationonScanNet
test mIoU· uses extra data· 2023-10-12
78.5
best: 79.8 (PTv3 ARKit LabelMaker)
SOTA
PonderV2: Pave the Way for 3D Foundation Model with A Universal Pre-training Paradigm arXiv:2310.08586
10-shot image generationonS3DIS
Mean IoU· uses extra data· 2023-10-12
79.9
best: 82.3 (Sonata + PTv3)
SOTA
PonderV2: Pave the Way for 3D Foundation Model with A Universal Pre-training Paradigm arXiv:2310.08586
10-shot image generationonS3DIS
oAcc· uses extra data· 2023-10-12
92.5
best: 93.3 (Sonata + PTv3)
SOTA
PonderV2: Pave the Way for 3D Foundation Model with A Universal Pre-training Paradigm arXiv:2310.08586
10-shot image generationonScanNet200
test mIoU· uses extra data· 2023-10-12
34.6
best: 44.9 (DITR)
SOTA
PonderV2: Pave the Way for 3D Foundation Model with A Universal Pre-training Paradigm arXiv:2310.08586
10-shot image generationonScanNet
val mIoU· uses extra data· 2023-10-12
77
best: 80.5 (DITR)
PonderV2: Pave the Way for 3D Foundation Model with A Universal Pre-training Paradigm arXiv:2310.08586
10-shot image generationonS3DIS Area5
mAcc· uses extra data· 2023-10-12
79
best: 81.6 (Sonata + PTv3)
PonderV2: Pave the Way for 3D Foundation Model with A Universal Pre-training Paradigm arXiv:2310.08586
10-shot image generationonS3DIS Area5
mIoU· uses extra data· 2023-10-12
73.2
best: 76 (Sonata + PTv3)
PonderV2: Pave the Way for 3D Foundation Model with A Universal Pre-training Paradigm arXiv:2310.08586
10-shot image generationonS3DIS Area5
oAcc· uses extra data· 2023-10-12
92.2
best: 93 (Sonata + PTv3)
PonderV2: Pave the Way for 3D Foundation Model with A Universal Pre-training Paradigm arXiv:2310.08586
10-shot image generationonS3DIS
mAcc· uses extra data· 2023-10-12
86.5
best: 89.9 (Sonata + PTv3)
PonderV2: Pave the Way for 3D Foundation Model with A Universal Pre-training Paradigm arXiv:2310.08586
10-shot image generationonScanNet200
val mIoU· uses extra data· 2023-10-12
32.3
best: 41.2 (DITR)
PonderV2: Pave the Way for 3D Foundation Model with A Universal Pre-training Paradigm arXiv:2310.08586

Computer Vision2 results

3D Semantic SegmentationonScanNet200
test mIoU· uses extra data· 2023-10-12
34.6
best: 44.9 (DITR)
SOTA
PonderV2: Pave the Way for 3D Foundation Model with A Universal Pre-training Paradigm arXiv:2310.08586
3D Semantic SegmentationonScanNet200
val mIoU· uses extra data· 2023-10-12
32.3
best: 41.2 (DITR)
PonderV2: Pave the Way for 3D Foundation Model with A Universal Pre-training Paradigm arXiv:2310.08586