{
  "id": "1406.6323",
  "version": "v1",
  "published": "2014-06-24T17:49:09.000Z",
  "updated": "2014-06-24T17:49:09.000Z",
  "title": "Dense Correspondences Across Scenes and Scales",
  "authors": [
    "Moria Tau",
    "Tal Hassner"
  ],
  "comment": "A longer version of this paper is in submission. Please see author homepage for an updated version",
  "categories": [
    "cs.CV"
  ],
  "abstract": "We seek a practical method for establishing dense correspondences between two images with similar content, but possibly different 3D scenes. One of the challenges in designing such a system is the local scale differences of objects appearing in the two images. Previous methods often considered only small subsets of image pixels; matching only pixels for which stable scales may be reliably estimated. More recently, others have considered dense correspondences, but with substantial costs associated with generating, storing and matching scale invariant descriptors. Our work here is motivated by the observation that pixels in the image have contexts -- the pixels around them -- which may be exploited in order to estimate local scales reliably and repeatably. Specifically, we make the following contributions. (i) We show that scales estimated in sparse interest points may be propagated to neighboring pixels where this information cannot be reliably determined. Doing so allows scale invariant descriptors to be extracted anywhere in the image, not just in detected interest points. (ii) We present three different means for propagating this information: using only the scales at detected interest points, using the underlying image information to guide the propagation of this information across each image, separately, and using both images simultaneously. Finally, (iii), we provide extensive results, both qualitative and quantitative, demonstrating that accurate dense correspondences can be obtained even between very different images, with little computational costs beyond those required by existing methods.",
  "revisions": [
    {
      "version": "v1",
      "updated": "2014-06-24T17:49:09.000Z"
    }
  ],
  "analyses": {
    "keywords": [
      "detected interest points",
      "information",
      "local scale differences",
      "sparse interest points",
      "estimate local scales"
    ],
    "note": {
      "typesetting": "TeX",
      "pages": 0,
      "language": "en",
      "license": "arXiv",
      "status": "editable",
      "adsabs": "2014arXiv1406.6323T"
    }
  }
}