1 / 21

CS 395/495-25: Spring 2004

CS 395/495-25: Spring 2004. IBMR: Poisson Solvers Can Reconstruct Images from their Changes Jack Tumblin jet@cs.northwestern.edu. Do pixels describe what we see?. What We Want. What We Get. What do you see?. A. B. What part has constant intensity?. What do you see?.

farica
Télécharger la présentation

CS 395/495-25: Spring 2004

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. CS 395/495-25: Spring 2004 IBMR:Poisson SolversCan Reconstruct Images from their Changes Jack Tumblin jet@cs.northwestern.edu

  2. Do pixels describe what we see? What We Want What We Get

  3. What do you see? A B What part has constant intensity?

  4. What do you see? Humans don’t sense intensities reliably, but infer them from changes A B B intensity is constant, A is darker on right

  5. What do you see? Humans don’t sense intensities reliably, but infer them from changes A B (tol’djah!)

  6. What do you see? X Y What part has constant intensity?

  7. What do you see? Humans don’t sense intensities reliably, but infer them from changes X Y What part has constant intensity? NEITHER!

  8. What do you see? Humans don’t sense intensities reliably, but infer them from changes X Y Constant What part has constant intensity? NEITHER!

  9. What do you see? Humans don’t sense intensities reliably, but infer them from changes Example: aren’t all the dots white? (http://udel.edu/~jgephart/fun2.htm)

  10. Why Pixels Could be Improved: • People see (or think they see) changesfinite features that may have infinite bandwidth occlusion, depth, collision time, trajectory changes, corner, cone tip, boundaries, edges, occlusions, shadow details, contact points, velocity & direction changes... Pixels only approximate changes, and approximate discontinuous changes poorly; object boundaries, silhouettes, etc. They force indirect estimation...

  11. How? Retinal Receptive Fields… - - - - - + - - - + + + + + + + + - • 130M Photoreceptors1M optic nerve fibers • Center-Surround Antagonism:Out  Center - (avg surround) • Complementary ON-center, OFF-center types • Center responds quickly; Surround responds more slowly • Output: ‘recent local change’

  12. Complementary Receptive Fields - - - - - + - - - Firing Rate (Hz) + 100 + + + + + 50 + + 10 ctr/surr - 10 0.1 1 ctr/surr 10 50 Firing Rate (Hz) 100 • Retina is ~differential for small signals • Better SNR • Can signal ambiguity (eyes closed, etc) • Allows quality/fault detection

  13. Yarbus (1950s): Pioneer of Retinal Stabilization Experiments (inspired a flood of others…)

  14. Strongly Implies ‘Filling In’ requires Nystagmus for temporal transients... ‘mm, nothing much. (green-ish?)’ ‘Not much to see.(pink-ish?)’ ‘BUT HERE is a Big ring of VERY strong change!’

  15. ‘mm, Not much to see. (green-ish?)’ ‘Not much to see.(pink-ish?)’

  16. What ‘Changes’ do we Sense? • Intensity (luminance) vs. local position • Color (chrominance) vs. local position • Intensity vs. time (‘flicker’) • Color vs. time • VERY weak, low-res: overall intensity • Inertial changes: movement, velocity… Compensated eye moves (saccade, glissade, smooth-pursuit… • Higher-level attributes? Umm, er, uh,….

  17. ‘Digital’ Image: a 2D Grid of Numbers • NO intrinsic meaning—use it for anything: reflectance, transparency, illumination, normal direction, material, velocity. BUT usually ‘intensity’ y y x x

  18. 2D Images Described by Change? • Image intensity as height field f(x,y): • 1st derivative— Gradient: the ‘uphill’ vector at point x,y = f(x,y) = (f(x,y)/x, f(x,y)/y) = f y x f(x,y)

  19. 2D Images Described by Change? • Image intensity as height field f(x,y): • 2st derivative— Gradient: the ‘uphill’ vector at point x,y = f(x,y) = (f(x,y)/x, f(x,y)/y) = f y x f(x,y)

  20. Review: Div, Grad and Curl • Formalized, computable ‘Local Change’

More Related