CSE152, Spr 07 Intro Computer VisionStereo Vision IIntroduction to Computer VisionCSE 152Lecture 13CSE152, Spr 07 Intro Computer VisionAnnouncements• Midterm• Out of • 78 pts.1824Low7596High4862Median52.67MeanScorePercentageCSE152, Spr 07 Intro Computer VisionShape-from-X(i.e., Reconstruction)• Methods for estimating 3-D shape from image data. X can be one (or more) of many cues.– Stereo (two or more views, known viewpoints)– Motion (moving camera or object)– Shading– Changing lighting (Photometric Stereo)– Texture variation– Focus/blur– Prior knowledge/context– structured light/lasersCSE152, Spr 07 Intro Computer VisionBinocular Stereopsis: MarsGiven two images of a scene where relative locations of cameras are known, estimate depth of all common scene points.Two images of MarsCSE152, Spr 07 Intro Computer VisionMar Rovers: Spirit and OpportunityFour pairsof stereo camerasCSE152, Spr 07 Intro Computer VisionAn Application: Mobile Robot NavigationThe Stanford Cart,H. Moravec, 1979.The INRIA Mobile Robot, 1990.Courtesy O. Faugeras and H. Moravec.CSE152, Spr 07 Intro Computer VisionCommercial Stereo HeadsTrinocular stereoTrinocular stereoBinocular stereoBinocular stereoCSE152, Spr 07 Intro Computer VisionStereo can work wellCSE152, Spr 07 Intro Computer VisionNeed for correspondenceTruco Fig. 7.5CSE152, Spr 07 Intro Computer VisionTriangulationNalwa Fig. 7.2CSE152, Spr 07 Intro Computer VisionStereo Vision Outline• Offline: Calibrate cameras & determine “epipolar geometry”• Online1. Acquire stereo images2. Rectify images to convenient epipolar geometry3. Establish correspondence 4. Estimate depthABCDCSE152, Spr 07 Intro Computer VisionBINOCULAR STEREO SYSTEMEstimating DepthZX(0,0) (d,0)Z=fXLXRDISPARITY(XL-XR)Z = (f/XL) XZ= (f/XR) (X-d)(f/XL) X = (f/XR) (X-d)X = (XLd) / (XL-XR)Z = d f(XL-XR)X = d XL(XL-XR)(Adapted from Hager)CSE152, Spr 07 Intro Computer VisionReconstruction: General 3-D case• Linear Method: find P such that• Non-Linear Method: find Q minimizingwhere q=MQ and q’=M’QGoal: Given two image measurements p and p’, estimate P.CSE152, Spr 07 Intro Computer VisionTwo Approaches1. Feature-Based– From each image, process “monocular” image to obtain image features or cues(e.g., corners, lines).– Establish correspondence between the detected features.2. Area-Based– Directly compare image regions between the two images.CSE152, Spr 07 Intro Computer VisionHuman Stereopsis: Binocular FusionHow are the correspondences established?Julesz (1971): Is the mechanism for binocular fusiona monocular process or a binocular one??• There is anecdotal evidence for the latter (camouflage).• Random dot stereograms provide an objective answerCSE152, Spr 07 Intro Computer VisionRandom Dot StereogramsCSE152, Spr 07 Intro Computer VisionRandom Dot
View Full Document
Unlocking...