TEACH VISION                                        David Young
                                                    January 1994

Suggested reading for an introductory course on Computer Vision.

-- Teach Files --------------------------------------------------------

The main sequence of teach files is TEACH VISION1 ... VISION7. To access
these files, you must first load LIB POPVISION.

-- Reading ------------------------------------------------------------

A note on mathematics:

    Sooner or later in vision work you will hit some mathematical
    formalism. Many of the authors below have found mathematics to be
    the most succinct way to express some of their arguments. However,
    most of the underlying ideas can be understood without recourse to
    mathematics, and the lectures will avoid its use as far as possible.
    Do not be discouraged by mathematical expressions in books; read
    round them, try to work out what is going on in general terms from
    the english text and diagrams, try different sources, and ask for an
    alternative explanation in seminars or tutorials.

Mathematics apart, some of these references assume a fair amount of
prior understanding - do not expect to be able to tackle everything. You
may find the full bibliography below helpful if you decide to do a
vision extended essay or project.

General Reading

    Sharples et al. (1989) is a quick and effective way into the topic.
    Winston (1981), Charniak & McDermott (1985), and Mayhew & Frisby
    (1984) all give broad accounts of AI approaches to vision. Marr
    (1982) is the source of many central ideas. Fischler & Firschein
    (1987b) is a good collection of important papers.  Boyle & Thomas
    (1988) is an introductory text which may be helpful for some
    technical aspects, particularly of low-level vision. Bruce & Green
    (1985) is an excellent introduction, but isn't particularly oriented
    towards AI/KBS. Sonka et al. (1993) is an up-to-date introduction
    mostly oriented towards low-level vision.

-- Bibliography -------------------------------------------------------

[Square-bracketed expressions are University of Sussex shelf marks.]

Ballard D.H. & Brown C.M. (1982). 'Computer Vision'. Englewood Cliffs,
    NJ: Prentice-Hall. [QZ 314 Bal]

    Generally a very technical presentation, with a quite liberal
    sprinkling of mathematics. A rich source of algorithms for all
    aspects of computer vision, though it is sometimes hard work to
    understand what they do and their limitations are not always brought
    out.

Barrow H.G. (1987). Learning receptive fields. Proceedings of the IEEE
    first annual international conference on neural networks, June 1987.
    [See David Young to borrow a copy.]

    Competitive learning produces some surprising results when applied
    to low-level vision.

Barrow H.G. & Tenenbaum J.M. (1981). Interpreting line drawings as
    three-dimensional surfaces. Artificial Intelligence, 17, 75-116.
    [QE 1 Art]. Reprinted in Brady (1981), pp. 75-116.

    Interesting introduction and discussion with a hard centre.

Blake A. & Yuille A. (1992). (Eds.) 'Active Vision'. Cambridge, MA: MIT
    Press. [TA 1632 Act]

Boden M.A. (1977). 'Artificial Intelligence and Natural Man'. Brighton:
    Harvester. [QZ 1240 Bod]

    Chapters 8 and 9 are largely a thorough run-down of the AI
    line-drawing analysis work up to 1977.

Boden M.A. (1988). 'Computer Models of Mind: Computational Approaches in
    Theoretical Psychology'. Cambridge: CUP. [QZ 1250 Bod]

    Chapters 2 and 3 give very interesting discussions of a variety of
    approaches.

Boyle R.D. & Thomas R.C. (1988). 'Computer Vision: A First Course'.
    Oxford: Blackwell Scientific. [TA 1632 Boy]

    A reasonably general introduction, though very sketchy in some
    areas.

Brady, M. (1981). (Ed.) 'Computer Vision'. Amsterdam: North-Holland.
    [QZ 1240 Art]

    Various important papers.

Brooks R.A. (1981). Symbolic reasoning among 3-D models and 2-D images.
    Artificial Intelligence, 17, 285-348. [QE 1 Art]. Reprinted in Brady
    (1981), pp. 285-348.

    A general philosophy of model-based vision and details (somewhat
    technical) of the ACRONYM system.

Brown C.M. (1988). (Ed.) 'Advances in Computer Vision' (2 Vols).
    Hillsdale, NJ: Lawrence Erlbaum. [QE 1 Adv]

    By no means a complete view of the field, but in interesting set of
    chapters by some important researchers; some present a slightly
    idiosyncratic viewpoint.

Bruce V. & Green P.R. (1985). 'Visual Perception: Physiology, Psychology
    and Ecology'. London: Lawrence Erlbaum. [QZ 314 Bru]

    An excellent general introduction, embracing and comparing all kinds
    of approaches, though not particularly oriented to AI methods. Very
    accessible.

Charniak E. & McDermott D. (1985). 'Introduction to Artificial
    Intelligence'. Reading, MA: Addison-Wesley. [QZ 1240 Cha]

    Chapter 3 is a brief and rather patchy but quite effective
    inroduction. Clear, mostly unmathematical.

Fischler M.A. & Firschein O. (1987a). 'The Eye, the Brain, and the
    Computer'. Reading, MA: Addison-Wesley. [QZ 1250 Fis]

    Beautifully presented and interesting text which has a much broader
    scope than computer vision.

Fischler M.A. & Firschein O. (1987b). (Eds.) 'Readings in Computer
    Vision: Issues, Problems, Principles and Paradigms'. Los Altos, CA:
    Morgan-Kaufmann. [Not yet in library.]

    Collection of significant papers.

Frisby J.P. (1979). 'Seeing: Illusion, Brain and Mind'. Oxford: Oxford
    University Press. [QZ 314 Fri]

    Very well presented introduction to the psychophysics and physiology
    of vision, with stimulating examples. Includes exciting spectacles.

Gibson J.J. (1966). 'The Senses Considered as Perceptual Systems'.
    Boston: Houghton Mifflin. [QZ 310 Gib]

Gibson J.J. (1979). 'The Ecological Approach to Visual Perception'.
    Boston: Houghton Mifflin. [QZ 314 Gib]

    People often love or hate Gibson's books. They are entirely
    unmathematical. They ignore AI completely. Many of his ideas are
    very stimulating.

Gonzalez R.C. & Wintz P. (1987). 'Digital Image Processing'. Reading,
    MA: Addison Wesley. [TA 1632 Gon]

    A fairly technical but comprehensive discussion of image processing
    techniques. Not particularly relevant to AI/KBS but lots of details
    of filtering methods and Fourier theory.

Gonzalez R.C. & Woods R.E. (1992). 'Digital Image Processing'. Reading,
    MA: Addison Wesley. [QE 1890 Gon]

    Updated version of Gonzalez & Wintz.

Haralick R.M. & Shapiro L.G. (1992). 'Computer and Robot Vision' (2
    Vols). Reading, MA: Addison-Wesley. [TA 1632 Har]

    Very full technical treatment, with many references and a lot of
    mathematical detail. Better structured than Ballard & Brown.

Hildreth E.C. (1984). Computations underlying the measurement of visual
    motion. Artificial Intelligence, 23, 309-354. [QE 1 Art]. Reprinted
    in Richards & Ullman (1987), pp. 99-146.

Hogg D. (1983). Model-based vision: a program to see a walking person.
    Image and Vision Computing, 1, 5-20. [See David Young for a copy]

    Description of WALKER.

Horn B.K.P. (1986). 'Robot Vision'. Cambridge MA: MIT Press.
    [TJ 211.3 Hor]

    A recent description of MIT work across the spectrum, but very
    mathematical.

Horn B.K.P. & Schunck B.G. (1981). Determining optical flow.
    Artificial Intelligence, 17, 185-204. [QE 1 Art]. Reprinted in
    Brady (1981), pp. 185-204.

    The smoothing approach to extracting optic flow. Mathematical, but
    maybe worth looking at the introduction and discussion.

Lee D.N. (1980). The optic flow field: the foundation of vision.
    Philosophical Transactions of the Royal Society of London, B 290,
    169-179. [QP 1 Phi]

    The use of optic flow information from the psychological point of
    view. Neglects the problem of determining the flow. Mostly
    accessible, some relatively simple mathematics.

Lee D.N. & Young D.S. (1985). Visual timing of interceptive action. In
    D.J. Ingle, M. Jeannerod & D.N. Lee (Eds.), 'Brain Mechanisms and
    Spatial Vision' (pp. 1-30). Dordrecht: Martinus Nijhoff.
    [QZ 314 Bra]

    How to use optic flow to control actions. Again assumes you can
    extract it, and inclined to muddle computational and algorithmic
    levels. Other papers in this volume may also be of interest.

Marr D. (1982). 'Vision: a Computational Investigation into the Human
    Representation and Processing of Visual Information'. San Francisco:
    W.H. Freeman. [QZ 1240 Mar]

    One of the most important recent books on vision. Mostly very
    readable. Very personal. Embraces AI, psychology and
    neurophysiology.

Mayhew J.E.W & Frisby J.P. (1981). Psychophysical and computational
    studies towards a theory of human stereopsis. Artificial
    Intelligence, 17, 349-386. [QE 1 Art]. Reprinted in Brady (1981),
    pp. 349-386.

    Psychophysics and computer modelling of stereopsis.

Mayhew J. & Frisby J. (1984). Computer vision. In T. O'Shea & M.
    Eisenstadt (Eds.), 'Artificial Intelligence: Tools, Techniques and
    Applications' (pp. 301-357). New York: Harper & Row. [QZ 1240 Art]

    A good, clear, general introduction, covering the main AI approaches
    in a critical and accessible way.

Murray D.W. (1987). Model-based recognition using 3D structure from
    motion. Image and Vision Computing, 5, 85-90. [See David Young to
    borrow a copy.]

    A powerful search method for model matching.

Nishihara H.K. (1981). Intensity, visible-surface, and volumetric
    representations. Artificial Intelligence, 17, 265-284. [QE 1 Art].
    Reprinted in Brady (1981), pp. 265-284.

    An alternative discussion of the Marr levels of understanding and
    representations.

Pentland A.P. (1986). (Ed.) 'From Pixels to Predicates: Recent Advances
    in Computational and Robotic Vision'. Norwood NJ: Ablex.
    [TA 1632 Fro]

    Some worthwhile papers in this collection.

Richards W. & Ullman S. (1987). (Eds.) 'Image Understanding 1985-86'.
    Norwood NJ: Ablex. [QZ 1390 Ima]

    A good collection of recent research.

Rumelhart D.E. & McClelland J.L. (1986). 'Parallel Distributed
    Processing: Explorations in the Microstructure of Cognition',
    Volume 1: 'Foundations'. Cambridge MA: MIT Press. [QZ 1000 Rum]

    An important source for this expanding field.

Sharples M., Hogg D., Hutchison C., Torrance S. & Young D. (1989).
    'Computers and Thought: An Introduction to Cognitive Science and
    Artificial Intelligence'. MIT Press - Bradford Books.

    Chapter 9 shows a variety of techniques applied to a very specific
    problem.

Sloman A. (1978). 'The Computer Revolution in Philosophy: Philosophy,
    Science and Models of Mind'. Hassocks, Sussex: Harvester.
    [QZ 1240 Slo]

    Chapter 9 treats vision as one aspect of the general problems of AI,
    with a few very specific examples.

Sonka M., Hlavac V. & Boyle R. (1993). 'Image Processing, Analysis and
    Machine Vision'. London: Chapman & Hall Computing. [TA 1632 Son]

    Good clear coverage of low-level vision and image processing.

Winston P.H. (1981). 'Artificial Intelligence' (2nd ed.). Reading, MA:
    Addison-Wesley. [QZ 1240 Win]

    Chapter 3 deals with the line-labelling technique; Chapter 10 gives
    a brief overview of AI vision methods.

--- $popvision/teach/vision
--- Copyright University of Sussex 1994. All rights reserved.
