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/OctFormer

OctFormer

Reported on 28 benchmarks across 9 tasks · 1 paper · 6 SOTA

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

Computer Vision8 results

3D Semantic SegmentationonScanNet200
test mIoU· 2023-05-04
32.5
best: 44.9 (DITR)
SOTA
OctFormer: Octree-based Transformers for 3D Point Clouds arXiv:2305.03045
3D Semantic SegmentationonScanNet200
val mIoU· 2023-05-04
32.6
best: 41.2 (DITR)
SOTA
OctFormer: Octree-based Transformers for 3D Point Clouds arXiv:2305.03045
Object DetectiononSUN-RGBD val
mAP@0.25· 2023-05-04
66.2
best: 69.7 (Point-GCC+TR3D+FF)
OctFormer: Octree-based Transformers for 3D Point Clouds arXiv:2305.03045
Object DetectiononSUN-RGBD val
mAP@0.5· 2023-05-04
50.6
best: 54 (Point-GCC+TR3D+FF)
OctFormer: Octree-based Transformers for 3D Point Clouds arXiv:2305.03045
3D Semantic SegmentationonScanNet++
Top-1 IoU· 2023-05-04
0.46
best: 0.525 (DITR)
OctFormer: Octree-based Transformers for 3D Point Clouds arXiv:2305.03045
3D Semantic SegmentationonScanNet++
Top-3 IoU· 2023-05-04
0.691
best: 0.762 (DITR)
OctFormer: Octree-based Transformers for 3D Point Clouds arXiv:2305.03045
3D Object DetectiononSUN-RGBD val
mAP@0.25· 2023-05-04
66.2
best: 69.7 (Point-GCC+TR3D+FF)
OctFormer: Octree-based Transformers for 3D Point Clouds arXiv:2305.03045
3D Object DetectiononSUN-RGBD val
mAP@0.5· 2023-05-04
50.6
best: 54 (Point-GCC+TR3D+FF)
OctFormer: Octree-based Transformers for 3D Point Clouds arXiv:2305.03045

Methodology8 results

3DonSUN-RGBD val
mAP@0.25· 2023-05-04
66.2
best: 69.7 (Point-GCC+TR3D+FF)
OctFormer: Octree-based Transformers for 3D Point Clouds arXiv:2305.03045
3DonSUN-RGBD val
mAP@0.5· 2023-05-04
50.6
best: 54 (Point-GCC+TR3D+FF)
OctFormer: Octree-based Transformers for 3D Point Clouds arXiv:2305.03045
2D ClassificationonSUN-RGBD val
mAP@0.25· 2023-05-04
66.2
best: 69.7 (Point-GCC+TR3D+FF)
OctFormer: Octree-based Transformers for 3D Point Clouds arXiv:2305.03045
2D ClassificationonSUN-RGBD val
mAP@0.5· 2023-05-04
50.6
best: 54 (Point-GCC+TR3D+FF)
OctFormer: Octree-based Transformers for 3D Point Clouds arXiv:2305.03045
2D Object DetectiononSUN-RGBD val
mAP@0.25· 2023-05-04
66.2
best: 69.7 (Point-GCC+TR3D+FF)
OctFormer: Octree-based Transformers for 3D Point Clouds arXiv:2305.03045
2D Object DetectiononSUN-RGBD val
mAP@0.5· 2023-05-04
50.6
best: 54 (Point-GCC+TR3D+FF)
OctFormer: Octree-based Transformers for 3D Point Clouds arXiv:2305.03045
16konSUN-RGBD val
mAP@0.25· 2023-05-04
66.2
best: 69.7 (Point-GCC+TR3D+FF)
OctFormer: Octree-based Transformers for 3D Point Clouds arXiv:2305.03045
16konSUN-RGBD val
mAP@0.5· 2023-05-04
50.6
best: 54 (Point-GCC+TR3D+FF)
OctFormer: Octree-based Transformers for 3D Point Clouds arXiv:2305.03045

Medical6 results

Semantic SegmentationonScanNet200
test mIoU· 2023-05-04
32.5
best: 44.9 (DITR)
SOTA
OctFormer: Octree-based Transformers for 3D Point Clouds arXiv:2305.03045
Semantic SegmentationonScanNet200
val mIoU· 2023-05-04
32.6
best: 41.2 (DITR)
SOTA
OctFormer: Octree-based Transformers for 3D Point Clouds arXiv:2305.03045
Semantic SegmentationonScanNet
test mIoU· 2023-05-04
76.6
best: 79.8 (PTv3 ARKit LabelMaker)
OctFormer: Octree-based Transformers for 3D Point Clouds arXiv:2305.03045
Semantic SegmentationonScanNet
val mIoU· 2023-05-04
75.7
best: 80.5 (DITR)
OctFormer: Octree-based Transformers for 3D Point Clouds arXiv:2305.03045
Semantic SegmentationonScanNet++
Top-1 IoU· 2023-05-04
0.46
best: 0.525 (DITR)
OctFormer: Octree-based Transformers for 3D Point Clouds arXiv:2305.03045
Semantic SegmentationonScanNet++
Top-3 IoU· 2023-05-04
0.691
best: 0.762 (DITR)
OctFormer: Octree-based Transformers for 3D Point Clouds arXiv:2305.03045

Audio6 results

10-shot image generationonScanNet200
test mIoU· 2023-05-04
32.5
best: 44.9 (DITR)
SOTA
OctFormer: Octree-based Transformers for 3D Point Clouds arXiv:2305.03045
10-shot image generationonScanNet200
val mIoU· 2023-05-04
32.6
best: 41.2 (DITR)
SOTA
OctFormer: Octree-based Transformers for 3D Point Clouds arXiv:2305.03045
10-shot image generationonScanNet
test mIoU· 2023-05-04
76.6
best: 79.8 (PTv3 ARKit LabelMaker)
OctFormer: Octree-based Transformers for 3D Point Clouds arXiv:2305.03045
10-shot image generationonScanNet
val mIoU· 2023-05-04
75.7
best: 80.5 (DITR)
OctFormer: Octree-based Transformers for 3D Point Clouds arXiv:2305.03045
10-shot image generationonScanNet++
Top-1 IoU· 2023-05-04
0.46
best: 0.525 (DITR)
OctFormer: Octree-based Transformers for 3D Point Clouds arXiv:2305.03045
10-shot image generationonScanNet++
Top-3 IoU· 2023-05-04
0.691
best: 0.762 (DITR)
OctFormer: Octree-based Transformers for 3D Point Clouds arXiv:2305.03045