Relations between the statistics of natural images and the response properties of cortical cells.

By David J. Field. 1987

How researchers see this paper

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

Citations in the corpus

Deep learning in neural networks: An overview (2014, background): “Many papers of the previous millennium, however, were about unsupervised learning (UL) without a teacher ( .g., Hebb, 1949; von der Malsburg, 1973; Kohonen, 1972, 1982, 1988; Willshaw and von der Malsburg, 19 76; Grossberg, 1976a,b; Watanabe, 1985; Pearlmutter and Hinton, 1986; Barrow, 1987; Field, ”
Measuring the tendency of CNNs to Learn Surface Statistical Regularities (2017, background): “” From [7, 4], we know that natural images tend to have the bulk of their Fourier spectrum concentrated on the low to mid range frequencies.”
Pros and Cons of GAN Evaluation Measures (2018, background): “The average power spectrum magnitude A over natural images has the form A(f) = 1/f−α, α ≈ 2 [17, 13, 9, 24].”
Could information theory provide an ecological theory of sensory processing? (2011, background): “…representation in the nervous system potentially has evolutionary advantages (Attneave 1954; Barlow 1961, 1985; Uttley 1979; Srinivisan et ai 1982; Field 1987; Linsker 1988, 1989a, b; Atick and Redlich 1990a, b; 1992a; Atick et al. 1990, 1991; Bialek et al. 1991b, see also Barlow 1989 and…”
Sparse Modeling for Image and Vision Processing (2014, methodology): “The dictionary learning formulation of Olshausen and Field [1996, 1997] is appropriate for processing natural image patches. A natural extension to full images instead of patches is called “convolutional sparse coding”. It consists of linearly decomposing an image by using small dictionary elements ”
Visual perception and the statistical properties of natural scenes. (2008, background): “In other words, for real images a given local sensor tends to respond relatively infrequently, but when it does respond it tends to produce a relatively large response (Field 1987, Olshausen & Field 1997).”
Criticality in the brain: A synthesis of neurobiology, models and cognition. (2017, background): “Close to the critical point, such fluctuations encompass a broad spectrum of scales, similar to the scale-invariant statistics of natural scenes (Field, 1987; Ruderman and Bialek, 1994).”
Efficient coding of natural sounds (2002, background): “Predicting Codes for Natural Sounds The notion of an efficient code cannot be separated from the ensemble of signals that are being encoded [6, 17].”
Hierarchical model of natural images and the origin of scale invariance (2013, background): “The scale invariance in natural images was ﬁ rst characterized by the 1/ f 2 spatial power spectrum of pixel intensities (2).”
Invariant Visual Object and Face Recognition: Neural and Computational Bases, and a Model, VisNet (2012, background): “The hypothesis thus is that the effects of redundancies in the input space of stimuli that result from the statistical properties of natural images (Field, 1987), together with the convergent architecture with competitive learning at each stage, produces a system that can perform invariant object re”
The Analysis of Image Contrast: From Quality Assessment to Automatic Enhancement (2016, methodology): “Here, Me n and M o n are evaluated using the logGabor filters [53].”
SAN: Learning Relationship between Convolutional Features for Multi-Scale Object Detection (2018, background): “Several researchers studied natural image statistics and the relationship between two resampled images [11,30,29].”