Abstract

In this paper, we establish a theoretical connection between the classical Lucas & Kanade (LK) algorithm and the emerging topic of Spatial Transformer Networks (STNs). STNs are of interest to the vision and learning communities due to their natural ability to combine alignment and classification within the same theoretical framework. Inspired by the Inverse Compositional (IC) variant of the LK algorithm, we present Inverse Compositional Spatial Transformer Networks (IC-STNs). We demonstrate that IC-STNs can achieve better performance than conventional STNs with less model capacity; in particular, we show superior performance in pure image alignment tasks as well as joint alignment/classification problems on real-world problems.

Video

Presentation & slides

The slides of the talk can be downloaded here.

Code

The code is hosted on GitHub (TensorFlow and PyTorch).

Publications

CVPR 2017 paper: [ link ]

arXiv preprint: https://arxiv.org/abs/1612.03897

BibTex:

@inproceedings{lin2017inverse,
  title={Inverse Compositional Spatial Transformer Networks},
  author={Lin, Chen-Hsuan and Lucey, Simon},
  booktitle={IEEE Conference on Computer Vision and Pattern Recognition ({CVPR})},
  year={2017}
}

Poster

(Click image to enlarge)