34
views
0
recommends
+1 Recommend
0 collections
    0
    shares
      • Record: found
      • Abstract: found
      • Article: found
      Is Open Access

      Wavelet Convolutional Neural Networks

      Preprint
      , ,

      Read this article at

      Bookmark
          There is no author summary for this article yet. Authors can add summaries to their articles on ScienceOpen to make them more accessible to a non-specialist audience.

          Abstract

          Spatial and spectral approaches are two major approaches for image processing tasks such as image classification and object recognition. Among many such algorithms, convolutional neural networks (CNNs) have recently achieved significant performance improvement in many challenging tasks. Since CNNs process images directly in the spatial domain, they are essentially spatial approaches. Given that spatial and spectral approaches are known to have different characteristics, it will be interesting to incorporate a spectral approach into CNNs. We propose a novel CNN architecture, wavelet CNNs, which combines a multiresolution analysis and CNNs into one model. Our insight is that a CNN can be viewed as a limited form of a multiresolution analysis. Based on this insight, we supplement missing parts of the multiresolution analysis via wavelet transform and integrate them as additional components in the entire architecture. Wavelet CNNs allow us to utilize spectral information which is mostly lost in conventional CNNs but useful in most image processing tasks. We evaluate the practical performance of wavelet CNNs on texture classification and image annotation. The experiments show that wavelet CNNs can achieve better accuracy in both tasks than existing models while having significantly fewer parameters than conventional CNNs.

          Related collections

          Most cited references14

          • Record: found
          • Abstract: not found
          • Article: not found

          Zur Theorie der orthogonalen Funktionensysteme

            Bookmark
            • Record: found
            • Abstract: found
            • Article: not found

            Texture classification and segmentation using wavelet frames.

            M. Unser (1995)
            This paper describes a new approach to the characterization of texture properties at multiple scales using the wavelet transform. The analysis uses an overcomplete wavelet decomposition, which yields a description that is translation invariant. It is shown that this representation constitutes a tight frame of l(2) and that it has a fast iterative algorithm. A texture is characterized by a set of channel variances estimated at the output of the corresponding filter bank. Classification experiments with l(2) Brodatz textures indicate that the discrete wavelet frame (DWF) approach is superior to a standard (critically sampled) wavelet transform feature extraction. These results also suggest that this approach should perform better than most traditional single resolution techniques (co-occurrences, local linear transform, and the like). A detailed comparison of the classification performance of various orthogonal and biorthogonal wavelet transforms is also provided. Finally, the DWF feature extraction technique is incorporated into a simple multicomponent texture segmentation algorithm, and some illustrative examples are presented.
              Bookmark
              • Record: found
              • Abstract: not found
              • Conference Proceedings: not found

              Deep filter banks for texture recognition and segmentation

                Bookmark

                Author and article information

                Journal
                20 May 2018
                Article
                1805.08620
                fd4ce537-0fe0-4b30-bc7c-31dd1243503c

                http://arxiv.org/licenses/nonexclusive-distrib/1.0/

                History
                Custom metadata
                10 pages, 7 figures, 5 tables. arXiv admin note: substantial text overlap with arXiv:1707.07394
                cs.CV cs.LG

                Computer vision & Pattern recognition,Artificial intelligence
                Computer vision & Pattern recognition, Artificial intelligence

                Comments

                Comment on this article