Today Perception using vision Range information from Vision Basic Image Processing Why is object recognition hard? -> “Computer Vision” with Jane Mulligan
Range sensing Last week Laser scanner (phase shift) Infrared (path loss) Ultrasound (time-of-flight) Today Depth from focus Depth from Stereo
Pin-Hole Camera A. Efros
Thin Lens Objects need to have the right distance to be in focus -> Depth-from-Focus method
Depth from Focus “in focus” + camera parameters = range How to test whether an image is “crisp” or “blurry”?
Testing for focus Unit Step -> 2nd Derivative Intuition: Images with high contrast have steep edges!
Convolution Calculate Laplacian / 2nd derivative by “convolving” image with 2D Kernel *
Depth from Stereo Distance between stereo pair known + distance in the image -> distance to object
Stereo Pairs Zero crossings of Laplacians of Gaussians Gaussians: blurred image (suppresses noise) Laplacians: edges Test how far similar edges are apart Epipolar constraints are given by the geometry of the Stereo pair
Other example for Convolutions: Canny Edge Detector 1. 2.+3. 4. Trace along ridges (non-maximum suppression) 15
Exercise: Thresholds 16 16 http://homepages.inf.ed.ac.uk/rbf/HIPR2/adpthrsh.htm Screen shots by Gary Bradski, 2005
Exercise: Morphological Operations Examples Morphology - applying Min-Max. Filters and its combinations Dilatation IB Opening IoB= (IB)B Erosion IB Image I Closing I•B= (IB)B TopHat(I)= I - (IB) BlackHat(I)= (IB) - I Grad(I)= (IB)-(IB)
Why is Object Recognition Hard?The difference between seeing and perception. Gary Bradski, 2009 19 What to do? Maybe we should try to find edges …. Gary Bradski, 2005
Surface orientation discontinuity
Reflectance discontinuity (i.e., change in surface material properties)
Illumination discontinuity (e.g., shadow)
Slide credit: Christopher Rasmussen 20 But, What’s an Edge?
To Deal With the Confusion, Your Brain has Rules...That can be wrong
We even see invisible edges…
And surfaces …
We need to deal with 3D Geometry 24 Perception is ambiguous … depending on your point of view! Graphic by Gary Bradski
And Lighting in 3D Which square is darker?
Lighting is Ill-posed … Perception of surfaces depends on lighting assumptions 26 Gary Bradski (c) 2008 26
Contrast 27 Which one is male and which one is female? Illusion by: Richard Russell,Harvard University Russell, R. (2009) A sex difference in facial pigmentation and its exaggeration by cosmetics. Perception, (38)1211-1219
Homework Read sections 4.2-5 (pages 145-180) Questionnaire on CU Learn Midterm: October 11 (during class)