COMP 546 - Computational Perception

I am offering the course in Winter 2024. Here is the Course Outline PDF. IMPORTANT: I will cover vision only, namely the lectures 1-14 from the 2018 materials below, plus new topics such as visual attention, natural scene statistics, neural network models of vision, and object & scene recognition. (Some of these topics were covered in 2019-2022 versions of the course, but I didn't add them to this public web page.)

LECTURE SCHEDULE (Winter 2018)

0. introduction (slides)
   levels of analysis in perception, course outline

1. geometry (slides) (notes)
   origins of spatial vision, visual angle, aperture, image projection, binocular disparity
2. focus and blur (slides) (notes)
   sampling, thin lens equation, depth of field, accommodation, aging and abnormal vision
3. photoreceptors, color (slides) (notes)
   spectra: emission, reflectance, absorptance; rods and cones, metamers, color displays, color blindness
4. retina (slides) (notes)
   spikes, color opponency, center-surround DOGs, cross-correlation
5. orientation selectivity (slides) (notes)
   retinotopic maps, simple cells, Gabor models
6. disparity tuned cells (slides) (notes)
   complex cells in V1, monocular vs. binocular, disparity space
7. image motion 1 (slides) (notes)
   XYT, time dependent receptive fields, 3D Gabors and sine waves, normal velocity
8. image motion 2 (slides) (notes)
   motion constraint equation, intersection of constraints, velocity tuned cells (MT)

9. egomotion (slides) (notes)
   translation and direction of heading; rotation: VOR, smooth pursuit eye movements
10. depth from blur, binocular steropsis (slides) (notes)
    blur on slanted planes, Panum's fusional area, accommodation-vergence conflict, random dot stereograms
11. shape from X: perspective, texture, shading (slides) (notes)
    vanishing points, depth gradient and texture cues, slant & tilt; curvature, Lambert's law
12. illumination and reflectance (slides) (notes)
    shape from shading (linear & cloudy day), lightness & color constancy

13. psychophysics (slides) (notes)
    psychometric curves, thresholds, contrast and disparity sensitivity
14. maximum likelihood (slides) (notes)
    examples of likelihoods, probability review
15. cue combinations, Bayesian models (slides) (notes)
    priors, MAP, depth reversal ambiguity

16. convolution (slides) (notes)
    impulse response functions, complex numbers review
17. Fourier transform and filtering (slides) (notes)
    examples, convolution theorem, low/band/high pass filters

18. sound 1 (slides) (notes)
    pressure vs. intensity, dB
19. sound 2 (slides) (notes)
    music and speech sounds, spectrograms

20. head and ear (slides) (notes)
    head and outer ear (HRIR, HRTF)
21. auditory pathway, sound localization (slides) (notes)
    cochlea and neural coding, duplex theory, Jeffress model, level and timing differences
22. auditory filters (slides) (notes)
    spectrograms revisited, critical bands and masking, spike triggered averaging and A1
23. echolocation and recognition by bats and porpoises (slides) (notes)
    constant frequency, frequency modulation, interference
24. What to review for final exam (slides)

EXERCISES