Effects of variability in synthetic training data on convolutional neural networks for 3D head reconstruction

@article{Gpfert2017EffectsOV,
  title={Effects of variability in synthetic training data on convolutional neural networks for 3D head reconstruction},
  author={Jan Philip G{\"o}pfert and Christina G{\"o}pfert and M. Botsch and B. Hammer},
  journal={2017 IEEE Symposium Series on Computational Intelligence (SSCI)},
  year={2017},
  pages={1-7}
}
  • Jan Philip Göpfert, Christina Göpfert, +1 author B. Hammer
  • Published 2017
  • Computer Science
  • 2017 IEEE Symposium Series on Computational Intelligence (SSCI)
  • Convolutional neural networks have recently shown great success in computer vision. They are able to automatically learn complicated mappings, often reaching human or super-human performance. However, a lack of labeled data can preclude the training of such networks. This is the case in the reconstruction of 3-dimensional human heads from 2-dimensional photographs. Approaching the problem backwards, starting from 3-dimensional heads and using photo-realistic rendering, one can create any number… CONTINUE READING

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 21 REFERENCES
    3D Face Reconstruction by Learning from Synthetic Data
    • 152
    • PDF
    Regressing Robust and Discriminative 3D Morphable Models with a Very Deep Neural Network
    • 224
    • PDF
    A convolutional neural network cascade for face detection
    • 825
    • PDF
    InverseFaceNet: Deep Single-Shot Inverse Face Rendering From A Single Image
    • 43
    • PDF
    Face recognition: a convolutional neural-network approach
    • 2,163
    • PDF
    A Neural Algorithm of Artistic Style
    • 1,277
    • PDF
    Deep Convolutional Network Cascade for Facial Point Detection
    • 1,020
    • PDF
    ImageNet classification with deep convolutional neural networks
    • 53,262
    • PDF