Empirically Analyzing the Effect of Dataset Biases on Deep Face Recognition Systems

@inproceedings{Kortylewski2018EmpiricallyAT,
  title={Empirically Analyzing the Effect of Dataset Biases on Deep Face Recognition Systems},
  author={Adam Kortylewski and Bernhard Egger and Andreas Schneider and Thomas Gerig and Andreas Morel-Forster and Thomas Vetter},
  booktitle={CVPR Workshops},
  year={2018}
}
In this document, we provide additional materials to supplement our main submission. We show that the generalization patterns which we observe for the AlexNet architecture, when biasing the training data to frontal faces as well as in the disentanglement experiment, can also be observed when training with the VGG-16 architecture. In particular, Figure 1 shows that the recognition rate of the biased VGG-16 network drops significantly for faces in an extreme yaw pose (red curve) compared to the… CONTINUE READING

From This Paper

Figures, tables, and topics from this paper.

Explore Further: Topics Discussed in This Paper

References

Publications referenced by this paper.
Showing 1-10 of 35 references

A Light CNN for Deep Face Representation With Noisy Labels

IEEE Transactions on Information Forensics and Security • 2018
View 2 Excerpts

Morphable Face Models - An Open Framework

2018 13th IEEE International Conference on Automatic Face & Gesture Recognition (FG 2018) • 2018
View 1 Excerpt

Scalismo Faces. https://github.com/unibas-gravis/ scalismo-faces/, 2016

A.F.B.E. Sandro Schoenborn, Andreas Schneider
[Online; accessed 01- • 2017
View 1 Excerpt

A comprehensive analysis of deep learning based representation for face recognition

M. Mehdipour Ghazi, H. Kemal Ekenel
Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops, pages 34– 41, • 2016
View 1 Excerpt

Similar Papers

Loading similar papers…