The “independent components” of natural scenes are edge filters

@article{Bell1997TheC,
  title={The “independent components” of natural scenes are edge filters},
  author={Anthony J. Bell and Terrence J. Sejnowski},
  journal={Vision Research},
  year={1997},
  volume={37},
  pages={3327-3338}
}
View on Elsevier
europepmc.org
2,417 Citations
Highly Influential Citations
196
Background Citations
924
Methods Citations
648
Results Citations
96

Figures from this paper

2,417 Citations

EXTENSIONS OF ICA AS MODELS OF NATURAL IMAGES AND VISUAL PROCESSING

Extensions of ICA are introduced that are based on modelling dependencies of the "independent" components estimated by basic ICA, and a multivariate autoregressive model of the dependencies lead to the concept of \double-blind" source separation.

MODELS OF IMAGES AND EARLY VISION

Extensions of ICA are introduced that are based on modelling dependencies of the ”independent” components estimated by basic ICA, which provides an alternative approach to the sparseness used in most models.

Statistical models of images and early vision

Work on modelling statistical regularities in ecologically valid visual input (“natural images”) and the obtained functional explanation of the properties of visual neurons are reviewed and ex-tensions of ICA are introduced based on modelling dependencies of the ”independent” components estimated by basic ICA.

Localized Receptive Fields May Mediate Transformation-Invariant Recognition in the Visual Cortex

It is shown that a relatively simple neural solution to the problem of transformation-invariant visual recognition also causes localized, oriented receptive fields to be learned from natural images.

Efficient coding of natural images with a population of noisy Linear-Nonlinear neurons

It is shown that an efficient coding model that incorporates biologically realistic ingredients - input and output noise, nonlinear response functions, and a metabolic cost on the firing rate - predicts receptive fields and response nonlinearities similar to those observed in the retina.

Statistical model of natural stimuli predicts edge-like pooling of spatial frequency channels in V2

The pooling that emerges allows the features to code for realistic low-level image features related to step edges in the modelled primary visual cortex to prove the viability of statistical modelling of natural images as a framework that produces quantitative predictions of visual processing.

Unsupervised Learning of Generative and Discriminative Weights Encoding Elementary Image Components in a Predictive Coding Model of Cortical Function

The proposed algorithm demonstrates for the first time that a single computational theory can explain the formation of cortical RFs and also the response properties of cortical neurons once those RFs have been learned.

The independent components of natural images are perceptually dependent

The pixel basis, the ICA basis and the discrete cosine basis are compared by asking subjects to interactively predict missing pixels (for the pixel basis) or to predict the coefficients of ICA and DCT basis functions in patches of natural images.

Development of localized oriented receptive fields by learning a translation-invariant code for natural images.

It is shown that a strategy for transformation-invariant coding of images based on a first-order Taylor series expansion of an image also causes localized, oriented receptive fields to be learned from natural image inputs.

Local statistics in natural scenes predict the saliency of synthetic textures

The results suggest that the principle of efficient coding not only accounts for filtering operations in the sensory periphery, but also shapes subsequent stages of sensory processing that are sensitive to high-order image statistics.
...

References

SHOWING 1-10 OF 56 REFERENCES

Natural image statistics and efficient coding.

It is suggested that a good objective for an efficient coding of natural Scenes is to maximize the sparseness of the representation, and it is shown that a network that learns sparse codes of natural scenes succeeds in developing localized, oriented, bandpass receptive fields similar to those in the mammalian striate cortex.

Natural Image Statistics and Eecient Coding

It is suggested that a good objective for an eecient coding of natural Scenes is to maximize the sparseness of the representation, and it is shown that a network that learns sparse codes of natural scenes succeeds in developing localized, oriented, bandpass receptive elds similar to those in the primate striate cortex.

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

  • D. Field
  • Computer Science
    Journal of the Optical Society of America. A, Optics and image science
  • 1987
The results obtained with six natural images suggest that the orientation and the spatial-frequency tuning of mammalian simple cells are well suited for coding the information in such images if the goal of the code is to convert higher-order redundancy into first- order redundancy.

Convergent Algorithm for Sensory Receptive Field Development

An unsupervised developmental algorithm for linear maps is derived which reduces the pixel-entropy at every update and thus removes pairwise correlations between pixels, and is biologically plausible since in a neural network implementation it requires only data available locally to a neuron.

Self-organization in a perceptual network

It is shown that even a single developing cell of a layered network exhibits a remarkable set of optimization properties that are closely related to issues in statistics, theoretical physics, adaptive signal processing, the formation of knowledge representation in artificial intelligence, and information theory.

Optimal Sampling of Natural Images: A Design Principle for the Visual System

The resulting optimal filters are remarkably similar to those observed in the mammalian visual cortex and the retinal ganglion cells of lower vertebrates.

Finding compact and sparse-distributed representations of visual images

An artificial neural network which self-organizes on the basis of simple Hebbian learning and negative feedback of activation is introduced and it is shown that it is capable both of forming compact codings of data distributions and of identifying filters most sensitive to sparse-distributed codes.

Searching for filters with 'interesting' output distributions: an uninteresting direction to explore?

It is argued that other constraints are required in order to understand the development of visual receptive fields and that filters can produce 'interesting' output distributions simply because natural images have variable local intensity variance.

Orientation selectivity of thalamic input to simple cells of cat visual cortex

It is reported that the orientation tuning of these potentials is almost unaffected by cooling the cortex, in agreement with Hubel and Wiesel's original proposal.

Optimal unsupervised learning in a single-layer linear feedforward neural network

...