1、Computational Vision,Jitendra MalikUniversity of California, Berkeley,What is in an image?,The input is just an array of brightness values; humans perceive structure in it.,From Pixels to Perception,outdoorwildlife,If visual processing was purely feedforward(it isnt),Pixels Local Neighborhoods,Bound
2、aries of image regions defined by a number of attributes,Brightness/colorTextureMotionBinocular disparityFamiliar configuration,Grouping is hierarchical,A,B,C,A,C are refinements of BA,C are mutual refinements A,B,C represent the same percept,Image,BG,L-bird,R-bird,grass,bush,head,eye,beak,far,body,
3、head,eye,beak,body,Perceptual organization forms a tree:,Two segmentations are consistent when they can beexplained by the samesegmentation tree,Humans assign a depth ordering to surfaces across a contour,R1 appears in front of R2R2 appears in front of R3,This can be done for images of natural scene
4、s ,Figure-Ground Labeling,- red is near; blue is far,Figure/Ground Organization,A contour belongs to one of the two (but not both) abutting regions.,Important for the perception of shape,Some other aspects of perceptual organization,Modal completion,What do we see here?,And here?,Some Pictorial Cues
5、,Support, Size,?,?,?,1,3,2,Cast Shadows,Shading,Measuring Surface Orientation,Binocular Stereopsis,Optical flow for a pilot,Object Category Recognition,Shape variation within a category,DArcy Thompson: On Growth and Form, 1917studied transformations between shapes of organisms,Attneaves Cat (1954)Li
6、ne drawings convey most of the information,Objects are in Scenes,Human stick figure from single image,Input image,Stick figure,Support masks,This is hard,Variety of posesClothingMissing partsSmall support for partsBackground clutter,Taxonomy and Partonomy,Taxonomy: E.g. Cats are in the order Felidae
7、 which in turn is in the class MammaliaRecognition can be at multiple levels of categorization, or be identification at the level of specific individuals , as in faces.Partonomy: Objects have parts, they have subparts and so on. The human body contains the head, which in turn contains the eyes.These
8、 notions apply equally well to scenes and to activities. Psychologists have argued that there is a “basic-level” at which categorization is fastest (Eleanor Rosch et al).In a partonomy each level contributes useful information for recognition.,Visual Control of Action,LocomotionNavigation/Way-findingObstacle AvoidanceManipulationGrasping Pick and PlaceTool use,Camera Obscura(Reinerus Gemma-Frisius, 1544),Camera Obscura(Angelo Sala, 1576-1637),
