The Devil Is in the Details: Delving Into Unbiased Data Processing for Human Pose Estimation

@article{Huang2020TheDI,
  title={The Devil Is in the Details: Delving Into Unbiased Data Processing for Human Pose Estimation},
  author={Jun-Jie Huang and Z. Zhu and Feng Guo and G. Huang},
  journal={2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
  year={2020},
  pages={5699-5708}
}
  • Jun-Jie Huang, Z. Zhu, +1 author G. Huang
  • Published 2020
  • Computer Science
  • 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
Recently, the leading performance of human pose estimation is dominated by top-down methods. Being a fundamental component in training and inference, data processing has not been systematically considered in pose estimation community, to the best of our knowledge. In this paper, we focus on this problem and find that the devil of top-down pose estimator is in the biased data processing. Specifically, by investigating the standard data processing in state-of-the-art approaches mainly including… Expand

Figures and Tables from this paper

TFPose: Direct Human Pose Estimation with Transformers
AID: Pushing the Performance Boundary of Human Pose Estimation with Information Dropping Augmentation
Deep Learning-Based Human Pose Estimation: A Survey
SIMPLE: SIngle-network with Mimicking and Point Learning for Bottom-up Human Pose Estimation
Recent Advances in Monocular 2D and 3D Human Pose Estimation: A Deep Learning Perspective
How to Train Your Robust Human Pose Estimator: Pay Attention to the Constraint Cue
Bottom-Up Human Pose Estimation Via Disentangled Keypoint Regression
Graph-PCNN: Two Stage Human Pose Estimation with Graph Pose Refinement
...
1
2
3
...

References

SHOWING 1-10 OF 37 REFERENCES
Learning Feature Pyramids for Human Pose Estimation
2D Human Pose Estimation: New Benchmark and State of the Art Analysis
PoseFix: Model-Agnostic General Human Pose Refinement Network
Towards Accurate Multi-person Pose Estimation in the Wild
RMPE: Regional Multi-person Pose Estimation
DeepCut: Joint Subset Partition and Labeling for Multi Person Pose Estimation
Pictorial structures revisited: People detection and articulated pose estimation
Deep High-Resolution Representation Learning for Human Pose Estimation
Realtime Multi-person 2D Pose Estimation Using Part Affinity Fields
...
1
2
3
4
...