Jointformer-flip

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

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

Computer Vision2 results

3D Human Pose EstimationonH3WB
MPJPE· 2022-08-07
63
best: 264.4 (CanonPose)
SOTA
Jointformer: Single-Frame Lifting Transformer with Error Prediction and Refinement for 3D Human Pose Estimation arXiv:2208.03704
Pose EstimationonH3WB
MPJPE· 2022-08-07
63
best: 264.4 (CanonPose)
SOTA
Jointformer: Single-Frame Lifting Transformer with Error Prediction and Refinement for 3D Human Pose Estimation arXiv:2208.03704

Methodology1 result

3DonH3WB
MPJPE· 2022-08-07
63
best: 264.4 (CanonPose)
SOTA
Jointformer: Single-Frame Lifting Transformer with Error Prediction and Refinement for 3D Human Pose Estimation arXiv:2208.03704

Audio1 result

1 Image, 2*2 StitchionH3WB
MPJPE· 2022-08-07
63
best: 264.4 (CanonPose)
SOTA
Jointformer: Single-Frame Lifting Transformer with Error Prediction and Refinement for 3D Human Pose Estimation arXiv:2208.03704