mlsanityData to 7 Oct 2026

The Devil is in the Details: Delving into Unbiased Data Processing for Human Pose Estimation

arXiv 1911.07524 ↗First posted 18 Nov 2019cs.CVcited by 79

Not on the map. Why?

Abstract (arXiv)

Being a fundamental component in training and inference, data processing has not been systematically considered in human pose estimation community, to the best of our knowledge. In this paper, we focus on this problem and find that the devil of human pose estimation evolution is in the biased data processing. Specifically, by investigating the standard data processing in state-of-the-art approaches mainly including coordinate system transformation and keypoint format transformation (i.e., encoding and decoding), we find that the results obtained by common flipping strategy are unaligned with the original ones in inference. Moreover, there is a statistical error in some keypoint format transformation methods. Two problems couple together, significantly degrade the pose estimation performance and thus lay a trap for the research community. This trap has given bone to many suboptimal remedies, which are always unreported, confusing but influential. By causing failure in reproduction and unfair in comparison, the unreported remedies seriously impedes the technological development. To tackle this dilemma from the source, we propose Unbiased Data Processing (UDP) consist of two technique aspect for the two aforementioned problems respectively (i.e., unbiased coordinate system transformation and unbiased keypoint format transformation). As a model-agnostic approach and a superior solution, UDP successfully pushes the performance boundary of human pose estimation and offers a higher and more reliable baseline for research community. Code is public available in https://github.com/HuangJunJie2017/UDP-Pose

Key facts. Cited by 79 papers in the map (data through 7 Oct 2026). First posted on arXiv 18 Nov 2019.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 16; 2022: 16; 2023: 19; 2024: 18; 2025: 7; 2026: 3.

79 papers.

Sort:

By area:

  1. 2021BEVDet: High-performance Multi-camera 3D Object Detection in Bird-Eye-Viewcited by 711
  2. 2022ViTPose: Simple Vision Transformer Baselines for Human Pose Estimationcited by 450
  3. 2023RTMPose: Real-Time Multi-Person Pose Estimation based on MMPosecited by 170
  4. 2021HRFormer: High-Resolution Transformer for Dense Predictioncited by 149
  5. 2022Revealing the Dark Secrets of Masked Image Modelingcited by 117
  6. 2021Bottom-Up Human Pose Estimation Via Disentangled Keypoint Regressioncited by 112
  7. 2021Pose Recognition with Cascade Transformerscited by 99
  8. 2021SimCC: a Simple Coordinate Classification Perspective for Human Pose Estimationcited by 68
  9. 2022ViTPose++: Vision Transformer for Generic Body Pose Estimationcited by 65
  10. 2021Deep Dual Consecutive Network for Human Pose Estimationcited by 51
  11. 2021Recent Advances in Monocular 2D and 3D Human Pose Estimation: A Deep Learning Perspectivecited by 49
  12. 2023Vision Transformer with Quadrangle Attentioncited by 46
38 more
  1. 2023Human Pose as Compositional Tokenscited by 45
  2. 2022Poseur: Direct Human Pose Regression with Transformerscited by 43
  3. 2021TFPose: Direct Human Pose Estimation with Transformerscited by 42
  4. 2023ViPLO: Vision Transformer based Pose-Conditioned Self-Loop Graph for Human-Object Interaction Detectioncited by 41
  5. 2021Polarized Self-Attention: Towards High-quality Pixel-wise Regressioncited by 39
  6. 2021Multi-Instance Pose Networks: Rethinking Top-Down Pose Estimationcited by 30
  7. 2022Temporal Feature Alignment and Mutual Information Maximization for Video-Based Human Pose Estimationcited by 24
  8. 20222D Human Pose Estimation: A Surveycited by 21
  9. 2023Vision-Based Human Pose Estimation via Deep Learning: A Surveycited by 20
  10. 2021The Center of Attention: Center-Keypoint Grouping via Attention for Multi-Person Pose Estimationcited by 18
  11. 2024RTMW: Real-Time Multi-Person 2D and 3D Whole-body Pose Estimationcited by 18
  12. 2022AggPose: Deep Aggregation Vision Transformer for Infant Pose Estimationcited by 13
  13. 2022QueryPose: Sparse Multi-Person Pose Regression via Spatial-Aware Part-Level Querycited by 13
  14. 2023Harnessing Diffusion Models for Visual Perception with Meta Promptscited by 13
  15. 2023DistilPose: Tokenized Pose Regression with Heatmap Distillationcited by 12
  16. 2023SHaRPose: Sparse High-Resolution Representation for Human Pose Estimationcited by 11
  17. 2021Low-resolution Human Pose Estimationcited by 10
  18. 2022Dite-HRNet: Dynamic Lightweight High-Resolution Network for Human Pose Estimationcited by 10
  19. 2022Explicit Occlusion Reasoning for Multi-person 3D Human Pose Estimationcited by 10
  20. 2023Semi-Supervised 2D Human Pose Estimation Driven by Position Inconsistency Pseudo Label Correction Modulecited by 10
  21. 2024Video-Based Human Pose Regression via Decoupled Space-Time Aggregationcited by 10
  22. 2022Heatmap Distribution Matching for Human Pose Estimationcited by 9
  23. 2024ProbPose: A Probabilistic Approach to 2D Human Pose Estimationcited by 9
  24. 2021Relation-Based Associative Joint Location for Human Pose Estimation in Videoscited by 7
  25. 2023Self-Correctable and Adaptable Inference for Generalizable Human Pose Estimationcited by 7
  26. 2023Action Recognition with Multi-stream Motion Modeling and Mutual Information Maximizationcited by 7
  27. 2023Lightweight Super-Resolution Head for Human Pose Estimationcited by 7
  28. 2022AdaptivePose++: A Powerful Single-Stage Network for Multi-Person Pose Regressioncited by 6
  29. 2023Revisiting Cephalometric Landmark Detection from the view of Human Pose Estimation with Lightweight Super-Resolution Headcited by 6
  30. 2023Human Pose-based Estimation, Tracking and Action Recognition with Deep Learning: A Surveycited by 6
  31. 2024Bridging the Domain Gap for Flight-Ready Spaceborne Visioncited by 6
  32. 2021SIMPLE: SIngle-network with Mimicking and Point Learning for Bottom-up Human Pose Estimationcited by 5
  33. 2023Human Body Pose Estimation for Gait Identification: A Comprehensive Survey of Datasets and Modelscited by 5
  34. 2023Learning Structure-Guided Diffusion Model for 2D Human Pose Estimationcited by 5
  35. 2022Self-Constrained Inference Optimization on Structural Groups for Human Pose Estimationcited by 4
  36. 2024PoseBench: Benchmarking the Robustness of Pose Estimation Models under Corruptionscited by 4
  37. 2023Bias-Compensated Integral Regression for Human Pose Estimationcited by 3
  38. 2024Benchmarking 2D Egocentric Hand Pose Datasetscited by 3

50 most cited of 79 All 79 (JSON)