Corpus ID: 232404369

TFPose: Direct Human Pose Estimation with Transformers

@article{Mao2021TFPoseDH,
  title={TFPose: Direct Human Pose Estimation with Transformers},
  author={Wei Mao and Yongtao Ge and Chunhua Shen and Zhi Tian and Xinlong Wang and Zhibin Wang},
  journal={ArXiv},
  year={2021},
  volume={abs/2103.15320}
}
We propose a human pose estimation framework that solves the task in the regression-based fashion. Unlike previous regression-based methods, which often fall behind those state-of-the-art methods, we formulate the pose estimation task into a sequence prediction problem that can effectively be solved by transformers. Our framework is simple and direct, bypassing the drawbacks of the heatmapbased pose estimation. Moreover, with the attention mechanism in transformers, our proposed framework is… Expand
Semi-Supervised 3D Hand-Object Poses Estimation with Interactions in Time
TransCrowd: Weakly-Supervised Crowd Counting with Transformer
Transformer Transforms Salient Object Detection and Camouflaged Object Detection
Temporal Action Proposal Generation with Transformers

References

SHOWING 1-10 OF 55 REFERENCES
Compositional Human Pose Regression
Human Pose Estimation with Spatial Contextual Information
DeepPose: Human Pose Estimation via Deep Neural Networks
Integral Human Pose Regression
HOT-Net: Non-Autoregressive Transformer for 3D Hand-Object Pose Estimation
Rethinking on Multi-Stage Networks for Human Pose Estimation
Towards Accurate Multi-person Pose Estimation in the Wild
2D Human Pose Estimation: New Benchmark and State of the Art Analysis
...
1
2
3
4
5
...