English

Fast Feature Extraction with CNNs with Pooling Layers

Computer Vision and Pattern Recognition 2018-05-09 v1 Machine Learning Machine Learning

Abstract

In 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.

Keywords

Cite

@article{arxiv.1805.03096,
  title  = {Fast Feature Extraction with CNNs with Pooling Layers},
  author = {Christian Bailer and Tewodros Habtegebrial and Kiran varanasi and Didier Stricker},
  journal= {arXiv preprint arXiv:1805.03096},
  year   = {2018}
}

Comments

Accepted at BMVC 2017

R2 v1 2026-06-23T01:48:35.332Z