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/TransFusion

TransFusion

Reported on 91 benchmarks across 12 tasks · 2 papers · 7 SOTA

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

Methodology49 results

Anomaly DetectiononMVTEC 3D-AD
Detection AUCROC· 2023-11-16
0.982
SOTA
TransFusion -- A Transparency-Based Diffusion Model for Anomaly Detection arXiv:2311.09999
3DonHumanEva-I
MMADE@2000ms· 2023-07-30
408
SOTA
TransFusion: A Practical and Effective Transformer-based Diffusion Model for 3D Human Motion Prediction arXiv:2307.16106
3DonHumanEva-I
MMFDE@2000ms· 2023-07-30
427
SOTA
TransFusion: A Practical and Effective Transformer-based Diffusion Model for 3D Human Motion Prediction arXiv:2307.16106
Anomaly DetectiononMVTec AD
Detection AUROC· uses extra data· 2023-11-16
99.4
best: 99.9 (GLASS)
TransFusion -- A Transparency-Based Diffusion Model for Anomaly Detection arXiv:2311.09999
Anomaly DetectiononMVTec AD
Segmentation AUPRO· uses extra data· 2023-11-16
95.3
best: 98.4 (WeakREST-Block)
TransFusion -- A Transparency-Based Diffusion Model for Anomaly Detection arXiv:2311.09999
Anomaly DetectiononVisA
Detection AUROC· 2023-11-16
98.7
best: 99.8 (UniNet)
TransFusion -- A Transparency-Based Diffusion Model for Anomaly Detection arXiv:2311.09999
Anomaly DetectiononVisA
Segmentation AUPRO (until 30% FPR)· 2023-11-16
94.7
best: 96.1 (AnomalyDINO-S (full-shot))
TransFusion -- A Transparency-Based Diffusion Model for Anomaly Detection arXiv:2311.09999
Anomaly DetectiononMVTEC 3D-AD
Segmentation AUPRO· 2023-11-16
0.983
best: 93.75 (CDO)
TransFusion -- A Transparency-Based Diffusion Model for Anomaly Detection arXiv:2311.09999
Anomaly DetectiononMVTEC 3D-AD
Detection AUROC· 2023-11-16
0.957
best: 93.7 (AST)
TransFusion -- A Transparency-Based Diffusion Model for Anomaly Detection arXiv:2311.09999
Anomaly DetectiononMVTEC 3D-AD
Segmentation AUPRO· 2023-11-16
0.947
best: 93.75 (CDO)
TransFusion -- A Transparency-Based Diffusion Model for Anomaly Detection arXiv:2311.09999
3DonAMASS
ADE· 2023-07-30
0.508
best: 0.656 (ThePoseKnows)
TransFusion: A Practical and Effective Transformer-based Diffusion Model for 3D Human Motion Prediction arXiv:2307.16106
3DonAMASS
APD· 2023-07-30
8.853
best: 13.17 (Dlow)
TransFusion: A Practical and Effective Transformer-based Diffusion Model for 3D Human Motion Prediction arXiv:2307.16106
3DonAMASS
FDE· 2023-07-30
0.568
best: 0.675 (ThePoseKnows)
TransFusion: A Practical and Effective Transformer-based Diffusion Model for 3D Human Motion Prediction arXiv:2307.16106
3DonHuman3.6M
ADE· 2023-07-30
358
best: 858 (HP-GAN)
TransFusion: A Practical and Effective Transformer-based Diffusion Model for 3D Human Motion Prediction arXiv:2307.16106
3DonHuman3.6M
APD· 2023-07-30
5975
best: 19466 (TCD)
TransFusion: A Practical and Effective Transformer-based Diffusion Model for 3D Human Motion Prediction arXiv:2307.16106
3DonHuman3.6M
FDE· 2023-07-30
468
best: 867 (HP-GAN)
TransFusion: A Practical and Effective Transformer-based Diffusion Model for 3D Human Motion Prediction arXiv:2307.16106
3DonHuman3.6M
MMADE· 2023-07-30
506
best: 847 (HP-GAN)
TransFusion: A Practical and Effective Transformer-based Diffusion Model for 3D Human Motion Prediction arXiv:2307.16106
3DonHuman3.6M
MMFDE· 2023-07-30
539
best: 883 (MT-VAE)
TransFusion: A Practical and Effective Transformer-based Diffusion Model for 3D Human Motion Prediction arXiv:2307.16106
3DonHumanEva-I
ADE@2000ms· 2023-07-30
204
best: 772 (HP-GAN)
TransFusion: A Practical and Effective Transformer-based Diffusion Model for 3D Human Motion Prediction arXiv:2307.16106
3DonHumanEva-I
APD@2000ms· 2023-07-30
1031
best: 6764 (TCD)
TransFusion: A Practical and Effective Transformer-based Diffusion Model for 3D Human Motion Prediction arXiv:2307.16106
3DonHumanEva-I
FDE@2000ms· 2023-07-30
234
best: 749 (HP-GAN)
TransFusion: A Practical and Effective Transformer-based Diffusion Model for 3D Human Motion Prediction arXiv:2307.16106
3DonnuScenes
NDS
0.72
best: 55.3 (LabelDistill)
3DonnuScenes
mAAE
0.13
best: 1 (BirdNet+ (multisweep))
3DonnuScenes
mAOE
0.36
best: 1.6 (PointNet)
3DonnuScenes
mAP
0.69
best: 45.1 (LabelDistill)
3DonnuScenes
mASE
0.24
best: 1 (qww)
3DonnuScenes
mATE
0.26
best: 1.06 (3D-GCK)
3DonnuScenes
mAVE
0.29
best: 2.21 (PointNet)
2D ClassificationonnuScenes
NDS
0.72
best: 55.3 (LabelDistill)
2D ClassificationonnuScenes
mAAE
0.13
best: 1 (BirdNet+ (multisweep))
2D ClassificationonnuScenes
mAOE
0.36
best: 1.6 (PointNet)
2D ClassificationonnuScenes
mAP
0.69
best: 45.1 (LabelDistill)
2D ClassificationonnuScenes
mASE
0.24
best: 1 (qww)
2D ClassificationonnuScenes
mATE
0.26
best: 1.06 (3D-GCK)
2D ClassificationonnuScenes
mAVE
0.29
best: 2.21 (PointNet)
2D Object DetectiononnuScenes
NDS
0.72
best: 55.3 (LabelDistill)
2D Object DetectiononnuScenes
mAAE
0.13
best: 1 (BirdNet+ (multisweep))
2D Object DetectiononnuScenes
mAOE
0.36
best: 1.6 (PointNet)
2D Object DetectiononnuScenes
mAP
0.69
best: 45.1 (LabelDistill)
2D Object DetectiononnuScenes
mASE
0.24
best: 1 (qww)
2D Object DetectiononnuScenes
mATE
0.26
best: 1.06 (3D-GCK)
2D Object DetectiononnuScenes
mAVE
0.29
best: 2.21 (PointNet)
16konnuScenes
NDS
0.72
best: 55.3 (LabelDistill)
16konnuScenes
mAAE
0.13
best: 1 (BirdNet+ (multisweep))
16konnuScenes
mAOE
0.36
best: 1.6 (PointNet)
16konnuScenes
mAP
0.69
best: 45.1 (LabelDistill)
16konnuScenes
mASE
0.24
best: 1 (qww)
16konnuScenes
mATE
0.26
best: 1.06 (3D-GCK)
16konnuScenes
mAVE
0.29
best: 2.21 (PointNet)

Computer Vision30 results

Pose EstimationonHumanEva-I
MMADE@2000ms· 2023-07-30
408
SOTA
TransFusion: A Practical and Effective Transformer-based Diffusion Model for 3D Human Motion Prediction arXiv:2307.16106
Pose EstimationonHumanEva-I
MMFDE@2000ms· 2023-07-30
427
SOTA
TransFusion: A Practical and Effective Transformer-based Diffusion Model for 3D Human Motion Prediction arXiv:2307.16106
Pose EstimationonAMASS
ADE· 2023-07-30
0.508
best: 0.656 (ThePoseKnows)
TransFusion: A Practical and Effective Transformer-based Diffusion Model for 3D Human Motion Prediction arXiv:2307.16106
Pose EstimationonAMASS
APD· 2023-07-30
8.853
best: 13.17 (Dlow)
TransFusion: A Practical and Effective Transformer-based Diffusion Model for 3D Human Motion Prediction arXiv:2307.16106
Pose EstimationonAMASS
FDE· 2023-07-30
0.568
best: 0.675 (ThePoseKnows)
TransFusion: A Practical and Effective Transformer-based Diffusion Model for 3D Human Motion Prediction arXiv:2307.16106
Pose EstimationonHuman3.6M
ADE· 2023-07-30
358
best: 858 (HP-GAN)
TransFusion: A Practical and Effective Transformer-based Diffusion Model for 3D Human Motion Prediction arXiv:2307.16106
Pose EstimationonHuman3.6M
APD· 2023-07-30
5975
best: 19466 (TCD)
TransFusion: A Practical and Effective Transformer-based Diffusion Model for 3D Human Motion Prediction arXiv:2307.16106
Pose EstimationonHuman3.6M
FDE· 2023-07-30
468
best: 867 (HP-GAN)
TransFusion: A Practical and Effective Transformer-based Diffusion Model for 3D Human Motion Prediction arXiv:2307.16106
Pose EstimationonHuman3.6M
MMADE· 2023-07-30
506
best: 847 (HP-GAN)
TransFusion: A Practical and Effective Transformer-based Diffusion Model for 3D Human Motion Prediction arXiv:2307.16106
Pose EstimationonHuman3.6M
MMFDE· 2023-07-30
539
best: 883 (MT-VAE)
TransFusion: A Practical and Effective Transformer-based Diffusion Model for 3D Human Motion Prediction arXiv:2307.16106
Pose EstimationonHumanEva-I
ADE@2000ms· 2023-07-30
204
best: 772 (HP-GAN)
TransFusion: A Practical and Effective Transformer-based Diffusion Model for 3D Human Motion Prediction arXiv:2307.16106
Pose EstimationonHumanEva-I
APD@2000ms· 2023-07-30
1031
best: 6764 (TCD)
TransFusion: A Practical and Effective Transformer-based Diffusion Model for 3D Human Motion Prediction arXiv:2307.16106
Pose EstimationonHumanEva-I
FDE@2000ms· 2023-07-30
234
best: 749 (HP-GAN)
TransFusion: A Practical and Effective Transformer-based Diffusion Model for 3D Human Motion Prediction arXiv:2307.16106
Multi-Object TrackingonnuScenes
AMOTA
0.72
best: 0.763 (MCTrack)
Object TrackingonnuScenes
AMOTA
0.72
best: 0.763 (MCTrack)
Object DetectiononnuScenes
NDS
0.72
best: 55.3 (LabelDistill)
Object DetectiononnuScenes
mAAE
0.13
best: 1 (BirdNet+ (multisweep))
Object DetectiononnuScenes
mAOE
0.36
best: 1.6 (PointNet)
Object DetectiononnuScenes
mAP
0.69
best: 45.1 (LabelDistill)
Object DetectiononnuScenes
mASE
0.24
best: 1 (qww)
Object DetectiononnuScenes
mATE
0.26
best: 1.06 (3D-GCK)
Object DetectiononnuScenes
mAVE
0.29
best: 2.21 (PointNet)
3D Object DetectiononnuScenes
NDS
0.72
best: 55.3 (LabelDistill)
3D Object DetectiononnuScenes
mAAE
0.13
best: 1 (BirdNet+ (multisweep))
3D Object DetectiononnuScenes
mAOE
0.36
best: 1.6 (PointNet)
3D Object DetectiononnuScenes
mAP
0.69
best: 45.1 (LabelDistill)
3D Object DetectiononnuScenes
mASE
0.24
best: 1 (qww)
3D Object DetectiononnuScenes
mATE
0.26
best: 1.06 (3D-GCK)
3D Object DetectiononnuScenes
mAVE
0.29
best: 2.21 (PointNet)
3D Multi-Object TrackingonnuScenes
AMOTA
0.72
best: 0.763 (MCTrack)

Audio13 results

1 Image, 2*2 StitchionHumanEva-I
MMADE@2000ms· 2023-07-30
408
SOTA
TransFusion: A Practical and Effective Transformer-based Diffusion Model for 3D Human Motion Prediction arXiv:2307.16106
1 Image, 2*2 StitchionHumanEva-I
MMFDE@2000ms· 2023-07-30
427
SOTA
TransFusion: A Practical and Effective Transformer-based Diffusion Model for 3D Human Motion Prediction arXiv:2307.16106
1 Image, 2*2 StitchionAMASS
ADE· 2023-07-30
0.508
best: 0.656 (ThePoseKnows)
TransFusion: A Practical and Effective Transformer-based Diffusion Model for 3D Human Motion Prediction arXiv:2307.16106
1 Image, 2*2 StitchionAMASS
APD· 2023-07-30
8.853
best: 13.17 (Dlow)
TransFusion: A Practical and Effective Transformer-based Diffusion Model for 3D Human Motion Prediction arXiv:2307.16106
1 Image, 2*2 StitchionAMASS
FDE· 2023-07-30
0.568
best: 0.675 (ThePoseKnows)
TransFusion: A Practical and Effective Transformer-based Diffusion Model for 3D Human Motion Prediction arXiv:2307.16106
1 Image, 2*2 StitchionHuman3.6M
ADE· 2023-07-30
358
best: 858 (HP-GAN)
TransFusion: A Practical and Effective Transformer-based Diffusion Model for 3D Human Motion Prediction arXiv:2307.16106
1 Image, 2*2 StitchionHuman3.6M
APD· 2023-07-30
5975
best: 19466 (TCD)
TransFusion: A Practical and Effective Transformer-based Diffusion Model for 3D Human Motion Prediction arXiv:2307.16106
1 Image, 2*2 StitchionHuman3.6M
FDE· 2023-07-30
468
best: 867 (HP-GAN)
TransFusion: A Practical and Effective Transformer-based Diffusion Model for 3D Human Motion Prediction arXiv:2307.16106
1 Image, 2*2 StitchionHuman3.6M
MMADE· 2023-07-30
506
best: 847 (HP-GAN)
TransFusion: A Practical and Effective Transformer-based Diffusion Model for 3D Human Motion Prediction arXiv:2307.16106
1 Image, 2*2 StitchionHuman3.6M
MMFDE· 2023-07-30
539
best: 883 (MT-VAE)
TransFusion: A Practical and Effective Transformer-based Diffusion Model for 3D Human Motion Prediction arXiv:2307.16106
1 Image, 2*2 StitchionHumanEva-I
ADE@2000ms· 2023-07-30
204
best: 772 (HP-GAN)
TransFusion: A Practical and Effective Transformer-based Diffusion Model for 3D Human Motion Prediction arXiv:2307.16106
1 Image, 2*2 StitchionHumanEva-I
APD@2000ms· 2023-07-30
1031
best: 6764 (TCD)
TransFusion: A Practical and Effective Transformer-based Diffusion Model for 3D Human Motion Prediction arXiv:2307.16106
1 Image, 2*2 StitchionHumanEva-I
FDE@2000ms· 2023-07-30
234
best: 749 (HP-GAN)
TransFusion: A Practical and Effective Transformer-based Diffusion Model for 3D Human Motion Prediction arXiv:2307.16106