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/SoftGroup

SoftGroup

Reported on 36 benchmarks across 8 tasks · 1 paper · 32 SOTA

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

Computer Vision28 results

Object DetectiononScanNetV2
mAP@0.25· 2022-03-03
71.6
best: 78.8 (DEST (based on V-DETR) (TTA))
SOTA
SoftGroup for 3D Instance Segmentation on Point Clouds arXiv:2203.01509
Object DetectiononScanNetV2
mAP@0.5· 2022-03-03
59.4
best: 67.9 (DEST (based on V-DETR) (TTA))
SOTA
SoftGroup for 3D Instance Segmentation on Point Clouds arXiv:2203.01509
Instance SegmentationonS3DIS
AP@50· uses extra data· 2022-03-03
68.9
best: 75.8 (OneFormer3D)
SOTA
SoftGroup for 3D Instance Segmentation on Point Clouds arXiv:2203.01509
Instance SegmentationonS3DIS
mAP· uses extra data· 2022-03-03
54.4
best: 64.5 (Mask3D)
SOTA
SoftGroup for 3D Instance Segmentation on Point Clouds arXiv:2203.01509
Instance SegmentationonS3DIS
mCov· uses extra data· 2022-03-03
69.3
best: 74.9 (ISBNet)
SOTA
SoftGroup for 3D Instance Segmentation on Point Clouds arXiv:2203.01509
Instance SegmentationonS3DIS
mPrec· uses extra data· 2022-03-03
75.3
best: 82.3 (OneFormer3D)
SOTA
SoftGroup for 3D Instance Segmentation on Point Clouds arXiv:2203.01509
Instance SegmentationonS3DIS
mWCov· uses extra data· 2022-03-03
71.7
best: 76.8 (ISBNet)
SOTA
SoftGroup for 3D Instance Segmentation on Point Clouds arXiv:2203.01509
Instance SegmentationonScanNet(v2)
mAP @ 50· 2022-03-03
76.1
best: 81.6 (Relation3D)
SOTA
SoftGroup for 3D Instance Segmentation on Point Clouds arXiv:2203.01509
Instance SegmentationonScanNet(v2)
mAP@25· 2022-03-03
86.5
best: 90.1 (Relation3D)
SOTA
SoftGroup for 3D Instance Segmentation on Point Clouds arXiv:2203.01509
Instance SegmentationonSTPLS3D
AP· 2022-03-03
47.3
best: 64.5 (EASE)
SOTA
SoftGroup for 3D Instance Segmentation on Point Clouds arXiv:2203.01509
Instance SegmentationonSTPLS3D
AP25· 2022-03-03
71.4
best: 86.9 (EASE)
SOTA
SoftGroup for 3D Instance Segmentation on Point Clouds arXiv:2203.01509
Instance SegmentationonSTPLS3D
AP50· 2022-03-03
63.1
best: 80.8 (EASE)
SOTA
SoftGroup for 3D Instance Segmentation on Point Clouds arXiv:2203.01509
3D Object DetectiononScanNetV2
mAP@0.25· 2022-03-03
71.6
best: 78.8 (DEST (based on V-DETR) (TTA))
SOTA
SoftGroup for 3D Instance Segmentation on Point Clouds arXiv:2203.01509
3D Object DetectiononScanNetV2
mAP@0.5· 2022-03-03
59.4
best: 67.9 (DEST (based on V-DETR) (TTA))
SOTA
SoftGroup for 3D Instance Segmentation on Point Clouds arXiv:2203.01509
3D Instance SegmentationonS3DIS
AP@50· uses extra data· 2022-03-03
68.9
best: 75.8 (OneFormer3D)
SOTA
SoftGroup for 3D Instance Segmentation on Point Clouds arXiv:2203.01509
3D Instance SegmentationonS3DIS
mAP· uses extra data· 2022-03-03
54.4
best: 64.5 (Mask3D)
SOTA
SoftGroup for 3D Instance Segmentation on Point Clouds arXiv:2203.01509
3D Instance SegmentationonS3DIS
mCov· uses extra data· 2022-03-03
69.3
best: 74.9 (ISBNet)
SOTA
SoftGroup for 3D Instance Segmentation on Point Clouds arXiv:2203.01509
3D Instance SegmentationonS3DIS
mPrec· uses extra data· 2022-03-03
75.3
best: 82.3 (OneFormer3D)
SOTA
SoftGroup for 3D Instance Segmentation on Point Clouds arXiv:2203.01509
3D Instance SegmentationonS3DIS
mWCov· uses extra data· 2022-03-03
71.7
best: 76.8 (ISBNet)
SOTA
SoftGroup for 3D Instance Segmentation on Point Clouds arXiv:2203.01509
3D Instance SegmentationonScanNet(v2)
mAP @ 50· 2022-03-03
76.1
best: 81.6 (Relation3D)
SOTA
SoftGroup for 3D Instance Segmentation on Point Clouds arXiv:2203.01509
3D Instance SegmentationonScanNet(v2)
mAP@25· 2022-03-03
86.5
best: 90.1 (Relation3D)
SOTA
SoftGroup for 3D Instance Segmentation on Point Clouds arXiv:2203.01509
3D Instance SegmentationonSTPLS3D
AP· 2022-03-03
47.3
best: 64.5 (EASE)
SOTA
SoftGroup for 3D Instance Segmentation on Point Clouds arXiv:2203.01509
3D Instance SegmentationonSTPLS3D
AP25· 2022-03-03
71.4
best: 86.9 (EASE)
SOTA
SoftGroup for 3D Instance Segmentation on Point Clouds arXiv:2203.01509
3D Instance SegmentationonSTPLS3D
AP50· 2022-03-03
63.1
best: 80.8 (EASE)
SOTA
SoftGroup for 3D Instance Segmentation on Point Clouds arXiv:2203.01509
Instance SegmentationonS3DIS
mRec· uses extra data· 2022-03-03
69.8
best: 77.1 (ISBNet)
SoftGroup for 3D Instance Segmentation on Point Clouds arXiv:2203.01509
Instance SegmentationonScanNet(v2)
mAP· 2022-03-03
50.4
best: 62.2 (Relation3D)
SoftGroup for 3D Instance Segmentation on Point Clouds arXiv:2203.01509
3D Instance SegmentationonS3DIS
mRec· uses extra data· 2022-03-03
69.8
best: 77.1 (ISBNet)
SoftGroup for 3D Instance Segmentation on Point Clouds arXiv:2203.01509
3D Instance SegmentationonScanNet(v2)
mAP· 2022-03-03
50.4
best: 62.2 (Relation3D)
SoftGroup for 3D Instance Segmentation on Point Clouds arXiv:2203.01509

Methodology8 results

3DonScanNetV2
mAP@0.25· 2022-03-03
71.6
best: 78.8 (DEST (based on V-DETR) (TTA))
SOTA
SoftGroup for 3D Instance Segmentation on Point Clouds arXiv:2203.01509
3DonScanNetV2
mAP@0.5· 2022-03-03
59.4
best: 67.9 (DEST (based on V-DETR) (TTA))
SOTA
SoftGroup for 3D Instance Segmentation on Point Clouds arXiv:2203.01509
2D ClassificationonScanNetV2
mAP@0.25· 2022-03-03
71.6
best: 78.8 (DEST (based on V-DETR) (TTA))
SOTA
SoftGroup for 3D Instance Segmentation on Point Clouds arXiv:2203.01509
2D ClassificationonScanNetV2
mAP@0.5· 2022-03-03
59.4
best: 67.9 (DEST (based on V-DETR) (TTA))
SOTA
SoftGroup for 3D Instance Segmentation on Point Clouds arXiv:2203.01509
2D Object DetectiononScanNetV2
mAP@0.25· 2022-03-03
71.6
best: 78.8 (DEST (based on V-DETR) (TTA))
SOTA
SoftGroup for 3D Instance Segmentation on Point Clouds arXiv:2203.01509
2D Object DetectiononScanNetV2
mAP@0.5· 2022-03-03
59.4
best: 67.9 (DEST (based on V-DETR) (TTA))
SOTA
SoftGroup for 3D Instance Segmentation on Point Clouds arXiv:2203.01509
16konScanNetV2
mAP@0.25· 2022-03-03
71.6
best: 78.8 (DEST (based on V-DETR) (TTA))
SOTA
SoftGroup for 3D Instance Segmentation on Point Clouds arXiv:2203.01509
16konScanNetV2
mAP@0.5· 2022-03-03
59.4
best: 67.9 (DEST (based on V-DETR) (TTA))
SOTA
SoftGroup for 3D Instance Segmentation on Point Clouds arXiv:2203.01509