Fast Dense Feature Extraction with CNNs that have Pooling or Striding Layers

Christian Bailer, Tewodros Amberbir Habtegebrial, Kiran Varanasi, Didier Stricker

In: BMVC. British Machine Vision Conference (BMVC-2017) September 4-7 London United Kingdom 2017.


n recent years, many publications showed that convolutional neural network based features can have a superior performance to engineered features. However, not much effort was taken so far to extract local features efficiently for a whole image. In this paper, we present an approach to compute patch-based local feature descriptors efficiently in presence of pooling and striding layers for whole images at once. Our approach is generic and can be applied to nearly all existing network architectures. This includes networks for all local feature extraction tasks like camera calibration, Patchmatching, optical flow estimation and stereo matching. In addition, our approach can be applied to other patch-based approaches like sliding window object detection and recognition. We complete our paper with a speed benchmark of popular CNN based feature extraction approaches applied on a whole image, with and without our speedup, and example code (for Torch) that shows how an arbitrary CNN architecture can be easily converted by our approach.


FastCNNFeature_BMVC.pdf (pdf, 239 KB) (zip, 4 KB)

Deutsches Forschungszentrum für Künstliche Intelligenz
German Research Center for Artificial Intelligence