DeepLabCut: markerless pose estimation of user-defined body parts with deep learning

Explore this paper's citation graph

Summary

This work presents an efficient method for markerless pose estimation based on transfer learning with deep neural networks that achieves excellent results with minimal training data and shows that the toolbox, called DeepLabCut, can achieve human accuracy with only a few hundred frames of training data.

Type
preprint
Published
2018-04-09
Cited by
4,558
References
82
Access
Open access

Keywords

Tracking (education), Artificial intelligence, Deep learning, Computer science, Computer vision

References

Cited by

Related papers