mlsanityData to 14 Oct 2026

Comparison of marker-less 2D image-based methods for infant pose estimation

arXiv 2410.04980 ↗First posted 7 Oct 2024cs.CVcited by 2

Not on the map. Why?

Abstract (arXiv)

In this study we compare the performance of available generic- and infant-pose estimators for a video-based automated general movement assessment (GMA), and the choice of viewing angle for optimal recordings, i.e., conventional diagonal view used in GMA vs. top-down view. We used 4500 annotated video-frames from 75 recordings of infant spontaneous motor functions from 4 to 26 weeks. To determine which pose estimation method and camera angle yield the best pose estimation accuracy on infants in a GMA related setting, the distance to human annotations and the percentage of correct key-points (PCK) were computed and compared. The results show that the best performing generic model trained on adults, ViTPose, also performs best on infants. We see no improvement from using infant-pose estimators over the generic pose estimators on our infant dataset. However, when retraining a generic model on our data, there is a significant improvement in pose estimation accuracy. The pose estimation accuracy obtained from the top-down view is significantly better than that obtained from the diagonal view, especially for the detection of the hip key-points. The results also indicate limited generalization capabilities of infant-pose estimators to other infant datasets, which hints that one should be careful when choosing infant pose estimators and using them on infant datasets which they were not trained on. While the standard GMA method uses a diagonal view for assessment, pose estimation accuracy significantly improves using a top-down view. This suggests that a top-down view should be included in recording setups for automated GMA research.

Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Builds on 19 of its 65 references that are in the map. First posted on arXiv 7 Oct 2024.

Builds on

19 of 65 references.

References · 19

  1. 2014Microsoft COCO: Common Objects in Contextcited by 17,608
  2. 2019Deep High-Resolution Representation Learning for Visual Recognitioncited by 1,420
  3. 2019Deep High-Resolution Representation Learning for Human Pose Estimationcited by 1,339
  4. 2018OpenPose: Realtime Multi-Person 2D Pose Estimation using Part Affinity Fieldscited by 1,162
  5. 2016Realtime Multi-Person 2D Pose Estimation using Part Affinity Fieldscited by 1,064
  6. 2022ViTPose: Simple Vision Transformer Baselines for Human Pose Estimationcited by 456
  7. 2016Convolutional Pose Machinescited by 325
  8. 2018Markerless tracking of user-defined features with deep learningcited by 206
  9. 2020BlazePose: On-device Real-time Body Pose trackingcited by 113
  10. 2020Invariant Representation Learning for Infant Pose Estimation with Small Datacited by 20
  11. 2022AggPose: Deep Aggregation Vision Transformer for Infant Pose Estimationcited by 13
  12. 2021A Spatio-temporal Attention-based Model for Infant Movement Assessment from Videoscited by 12
7 more
  1. 2020Preterm infants' pose estimation with spatio-temporal featurescited by 9
  2. 2023Efficient Domain Adaptation via Generative Prior for 3D Infant Pose Estimationcited by 8
  3. 2020EfficientPose: Scalable single-person pose estimationcited by 7
  4. 2024Automatic infant 2D pose estimation from videos: comparing seven deep neural network methodscited by 6
  5. 2024Deep learning empowered sensor fusion boosts infant movement classificationcited by 4
  6. 2022Open video data sharing in developmental and behavioural sciencecited by 3
  7. 2023Mobile solutions for clinical surveillance and evaluation in infancy -- General Movement Appscited by 2

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 2.

2 papers.

  1. 2026Markerless Motion Capture for Biomechanical Whole-Body Kinematic Estimation in Infantscited by 1
  2. 2026The Blind Spot in 2D Infants' Pose Estimation:Robust Learning from Noisy Annotationscited by 1