Corpus ID: 212628502

Metric-Scale Truncation-Robust Heatmaps for 3D Human Pose Estimation

@article{Srndi2020MetricScaleTH,
  title={Metric-Scale Truncation-Robust Heatmaps for 3D Human Pose Estimation},
  author={Istv{\'a}n S{\'a}r{\'a}ndi and Timm Linder and Kai Oliver Arras and Bastian Leibe},
  journal={ArXiv},
  year={2020},
  volume={abs/2003.02953}
}
  • István Sárándi, Timm Linder, +1 author Bastian Leibe
  • Published in ArXiv 2020
  • Computer Science
  • Heatmap representations have formed the basis of 2D human pose estimation systems for many years, but their generalizations for 3D pose have only recently been considered. This includes 2.5D volumetric heatmaps, whose X and Y axes correspond to image space and the Z axis to metric depth around the subject. To obtain metric-scale predictions, these methods must include a separate, explicit post-processing step to resolve scale ambiguity. Further, they cannot encode body joint positions outside… CONTINUE READING

    Citations

    Publications citing this paper.

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 67 REFERENCES

    Latent structured models for human pose estimation

    VIEW 10 EXCERPTS
    HIGHLY INFLUENTIAL

    Compositional Human Pose Regression

    VIEW 4 EXCERPTS
    HIGHLY INFLUENTIAL

    Integral Human Pose Regression

    VIEW 12 EXCERPTS
    HIGHLY INFLUENTIAL

    Coarse-to-Fine Volumetric Prediction for Single-Image 3D Human Pose

    VIEW 7 EXCERPTS
    HIGHLY INFLUENTIAL

    Monocular 3D Human Pose Estimation in the Wild Using Improved CNN Supervision

    VIEW 6 EXCERPTS
    HIGHLY INFLUENTIAL

    3D Human Pose Estimation With 2D Marginal Heatmaps

    VIEW 5 EXCERPTS
    HIGHLY INFLUENTIAL

    A Simple Yet Effective Baseline for 3d Human Pose Estimation

    VIEW 4 EXCERPTS
    HIGHLY INFLUENTIAL

    Towards 3D Human Pose Estimation in the Wild: A Weakly-Supervised Approach

    VIEW 5 EXCERPTS
    HIGHLY INFLUENTIAL