The Structure of Locally Orderless Images

By Jan J. Koenderink, Andrea J. Koenderink-van Doorn. 1999

Subfields: Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

Citations in the corpus

Scaling the Scattering Transform: Deep Hybrid Networks (2017, neutral): “Indeed, many works in classical image processing or classification [18, 4, 30, 28] suggests that the local encoding of an image permit to describe efficiently an image.”
Beyond Bags of Features: Spatial Pyramid Matching for Recognizing Natural Scene Categories (2006, background): “Koen-derink and Van Doorn [10] have generalized histograms to locally orderless images , or histogram-valued scale spaces (i.e., for each Gaussian aperture at a given location and scale, the locally orderless image returns the histogram of image features aggregated over that aperture).”
Spatial pyramid matching (2009, unknown): “A possible unifying theory underlying the seemingly disparate variety of subdivide-and-disorder techniques in the literature is oﬀered by the concept of locally orderless images of Koenderink and Van Doorn (1999).”
Ask the locals: Multi-way local pooling for image recognition (2011, background): “It is related to Koenderink’s concept of locally orderless images [24], and can be traced back to Hubel and Wiesel’s seminal work on complex cells in the visual cortex [18].”
Fundamental relationship between bilateral filtering, adaptive smoothing, and the nonlinear diffusion equation (2002, methodology): “268, pp. 1417-1420, 1995. [ 5 ] F.R. Chen and D.S.”
A sparse texture representation using local affine regions (2005, background): “advocated by Koenderink and Van Doorn [19] because it alleviates aliasing effects.”
Learnable Pooling Regions for Image Classification (2013, methodology): “Motivated from biology [Riesenhuber and Poggio, 2009] and statistics of locally orderless images [Koenderink and Van Doorn, 1999], the spatial pooling approach has been found useful as an intermediate step of many today’s computer vision methods.”
Active shape model segmentation with optimal features (2002, methodology): “The size of this window function is characterized by a second scale parameter . The construction of local histograms, extracted from a Gaussian aperture function, is called a locally orderless image and discussed in [ 36 ].”
An Affine Invariant Salient Region Detector (2004, background): “Griffin [6] and Kœnderink and van Doorn [12] have suggested weighting histogram counts using a Gaussian window, but not in relation to anti-aliasing.”
A Fast Approximation of the Bilateral Filter Using a Signal Processing Approach (2006, background): “Koenderink and van Doorn (1999 ) h ave shown that local histograms are a useful tool to study image structure and content.”
Learning Semantic Image Representations at a Large Scale (2014, background): “The spatial invariance property also reflects the concept of locally orderless images [63], which suggests that low-level features are grouped spatially to provide information about the overall semantics.”
Locally Orderless Tracking (2012, methodology): “We are inspired by the work of Koenderink and Van Doorn on the structure of locally orderless images [1] which proposes an image representation method where the amount of spatial order preserved globally and locally can be tuned using two parameters.”
Learning Smooth Pooling Regions for Visual Recognition (2013, methodology): “Motivated from biology [7] [9] [19] [25] and statistics of locally orderless images [12], the spatial pooling approach has been found useful as an intermediate step of many today's computer vision methods ranging from local features based approaches [16] [30] to higherlevel semantic representations…”