Chained Representation Cycling: Learning to Estimate 3D Human Pose and Shape by Cycling Between Representations

@inproceedings{Regg2020ChainedRC,
  title={Chained Representation Cycling: Learning to Estimate 3D Human Pose and Shape by Cycling Between Representations},
  author={N. R{\"u}egg and Christoph Lassner and Michael J. Black and K. Schindler},
  booktitle={AAAI},
  year={2020}
}
  • N. Rüegg, Christoph Lassner, +1 author K. Schindler
  • Published in AAAI 2020
  • Computer Science
  • The goal of many computer vision systems is to transform image pixels into 3D representations. Recent popular models use neural networks to regress directly from pixels to 3D object parameters. Such an approach works well when supervision is available, but in problems like human pose and shape estimation, it is difficult to obtain natural images with 3D ground truth. To go one step further, we propose a new architecture that facilitates unsupervised, or lightly supervised, learning. The idea is… CONTINUE READING

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 45 REFERENCES
    Unsupervised Geometry-Aware Representation for 3D Human Pose Estimation
    79
    Unite the People: Closing the Loop Between 3D and 2D Human Representations
    184
    End-to-End Recovery of Human Shape and Pose
    422
    Keep It SMPL: Automatic Estimation of 3D Human Pose and Shape from a Single Image
    515
    Coarse-to-Fine Volumetric Prediction for Single-Image 3D Human Pose
    371
    BodyNet: Volumetric Inference of 3D Human Body Shapes
    130
    A Simple Yet Effective Baseline for 3d Human Pose Estimation
    387
    Learning Human Pose from Unaligned Data through Image Translation
    1
    Ordinal Depth Supervision for 3D Human Pose Estimation
    115