Stanford CS 223 B Computer Vision Winter 2006
Stanford CS 223 B Computer Vision, Winter 2006 Lecture 1 Intro and Image Formation Professor Sebastian Thrun CAs: Dan Maynes-Aminzade and Mitul Saha Guest lectures: Rick Szeliski (Microsoft Research) and Gary Bradski (Intel Research and Stanford) Sebastian Thrun CS 223 B Computer Vision, Winter 2006 1
Today’s Goals • Learn about CS 223 b • Get Excited about Computer Vision • Learn about Image Formation (Part 1) Sebastian Thrun CS 223 B Computer Vision, Winter 2006 2
Administrativa • Time and Location Mon/Wed 11: 00 -12: 20, Mc. Cullough 115 SCPD Televised • Web site http: //cs 223 b. stanford. edu Sebastian Thrun CS 223 B Computer Vision, Winter 2006 3
People Involved • You: 90 students signed up • Me: Sebastian Thrun – Office hours Wed 3 -4 with appointment, Gates 154 • CA: Mitul Saha – office hours Tue, Thu 3 -5 pm, Clark S 264 • CA: Dan Maynes-Aminzade – office hours Mon, Wed 3 -5 pm, Gates 386 • Guest lectureres – Gary Bradski, Intel Research and Stanford – Rick Szeliski, Microsoft Research Sebastian Thrun CS 223 B Computer Vision, Winter 2006 4
Guest Lecturers Sebastian Thrun CS 223 B Computer Vision, Winter 2006 5
Goals • To familiarize you with basic the techniques and jargon in the field • To enable you to solve computer vision problems • To let you experience (and appreciate!) the difficulties of real-world computer vision • To get you excited! Sebastian Thrun CS 223 B Computer Vision, Winter 2006 6
Course Requirements + Criteria • You have to – Turn in all assignments (30% of final grade) – Pass the midterm (30%) – Pass the competition (40%) • Late policy – Six late days • Teaming: – Assignments/competition: up to three students Sebastian Thrun CS 223 B Computer Vision, Winter 2006 7
Course Overview • Basics – Image Formation and Camera Calibration – Image Features – Calibration • 3 D Reconstruction – Stereo – Image Mosaics, Stiching • Motion – Optical Flow – Structure From Motion – Tracking • Object detection and recognition – – Grouping Detection Segmentation Classification Sebastian Thrun CS 223 B Computer Vision, Winter 2006 8
Course Overview Sebastian Thrun CS 223 B Computer Vision, Winter 2006 9
The Text Sebastian Thrun CS 223 B Computer Vision, Winter 2006 10
The Class Competition • No Major Project • Instead: A competition Sebastian Thrun CS 223 B Computer Vision, Winter 2006 11
The Competition: Motivation Sebastian Thrun CS 223 B Computer Vision, Winter 2006 12
Implications • Why not run a competition in CS 223 b? Sebastian Thrun CS 223 B Computer Vision, Winter 2006 13
The Competition • March 13, 11 -11: 30 am: The Competition – Given a stream of images acquired by a vehicle on a highway – Predict a classification of moving/non moving objects 5 seconds ahead • Same data used in all programming assignments – HW 1: Feature/object detection (due Jan 23) – HW 2: Camera calibration (due Jan 30) – HW 3: Visual odometry (due Feb 13) Sebastian Thrun CS 223 B Computer Vision, Winter 2006 14
The Competition, Example This is not the real data. We’ll collect the data with Stanley Sebastian Thrun CS 223 B Computer Vision, Winter 2006 15
Today’s Goals • Learn about CS 223 b • Get Excited about Computer Vision • Learn about image formation (Part 1) Sebastian Thrun CS 223 B Computer Vision, Winter 2006 16
Computer Graphics Output Image Synthetic Camera Model (slides courtesy of Michael Cohen) Sebastian Thrun CS 223 B Computer Vision, Winter 2006 17
Computer Vision Output Model Real Scene Real Cameras (slides courtesy of Michael Cohen) Sebastian Thrun CS 223 B Computer Vision, Winter 2006 18
Combined Output Image Synthetic Camera Model Real Scene Real Cameras (slides courtesy of Michael Cohen) Sebastian Thrun CS 223 B Computer Vision, Winter 2006 19
Example 1: Stereo See http: //schwehr. org/photo. Real. VR/example. html Sebastian Thrun CS 223 B Computer Vision, Winter 2006 20
Example 2: Structure From Motion http: //medic. rad. jhmi. edu/pbazin/perso/Research/Sf. Mvideo. html Sebastian Thrun CS 223 B Computer Vision, Winter 2006 21
Example 3: 3 D Modeling http: //www. photogrammetry. ethz. ch/research/cause/3 dreconstruction 3. html Sebastian Thrun CS 223 B Computer Vision, Winter 2006 22
Example 4: 3 D Modeling Drago Anguelov Sebastian Thrun CS 223 B Computer Vision, Winter 2006 23
Example 4: 3 D Modeling Sebastian Thrun CS 223 B Computer Vision, Winter 2006 24
Example 4: 3 D Modeling Sebastian Thrun CS 223 B Computer Vision, Winter 2006 25
Example 5: Segmentation http: //elib. cs. berkeley. edu/photos/classify/ Sebastian Thrun CS 223 B Computer Vision, Winter 2006 26
Example 6: Classification Sebastian Thrun CS 223 B Computer Vision, Winter 2006 27
Example 6: Classification Sebastian Thrun CS 223 B Computer Vision, Winter 2006 28
Example 7: Optical Flow Demo: Dirt Road Andrew Lookingbill, David Lieb, CS 223 b Winter 2004 Sebastian Thrun CS 223 B Computer Vision, Winter 2006 29
Example 8: Detection David Stavens, Andrew Lookingbill, David Lieb, CS 223 b Winter 2004 Sebastian Thrun CS 223 B Computer Vision, Winter 2006 30
Example 9: Tracking http: //www. seeingmachines. com/facelab. htm Sebastian Thrun CS 223 B Computer Vision, Winter 2006 31
Example 10: Human Vision Sebastian Thrun CS 223 B Computer Vision, Winter 2006 32
Example 9: Human Vision Sebastian Thrun CS 223 B Computer Vision, Winter 2006 33
Excited Yet? Sebastian Thrun CS 223 B Computer Vision, Winter 2006 34
Today’s Goals • Learn about CS 223 b • Get Excited about Computer Vision • Learn about image formation (Part 1) Sebastian Thrun CS 223 B Computer Vision, Winter 2006 35
Topics • • Pinhole Camera Orthographic Projection Perspective Camera Model Weak-Perspective Camera Model Sebastian Thrun CS 223 B Computer Vision, Winter 2006 36
Pinhole Camera -- Brunelleschi, XVth Century *many slides in this lecture from Marc Pollefeys comp 256, Lect 2 Sebastian Thrun CS 223 B Computer Vision, Winter 2006 37
Perspective Projection A “similar triangle’s” approach to vision. Sebastian Thrun CS 223 B Computer Vision, Winter 2006 Marc 38 Pollefeys
Perspective Projection X O x -x f Z Sebastian Thrun f CS 223 B Computer Vision, Winter 2006 39
Consequences: Parallel lines meet • There exist vanishing points Marc Pollefeys Sebastian Thrun CS 223 B Computer Vision, Winter 2006 40
The Effect of Perspective Sebastian Thrun CS 223 B Computer Vision, Winter 2006 41
Vanishing points H VPL VPR VP 2 VP 1 Different directions correspond to different vanishing points Sebastian Thrun VP 3 CS 223 B Computer Vision, Winter 2006 Marc Pollefeys 42
Implications For Perception* Same size things get smaller, we hardly notice… Parallel lines meet at a point… * A Cartoon Epistemology: http: //cns-alumni. bu. edu/~slehar/cartoonepist. html Sebastian Thrun CS 223 B Computer Vision, Winter 2006 43
Perspective Projection O X Z Sebastian Thrun -x f CS 223 B Computer Vision, Winter 2006 44
Weak Perspective Projection Z O Z Sebastian Thrun Z -x f CS 223 B Computer Vision, Winter 2006 45
Generalization of Orthographic Projection When the camera is at a (roughly constant) distance from the scene, take m=1. Sebastian Thrun CS 223 B Computer Vision, Winter 2006 Marc 46 Pollefeys
Pictorial Comparison Weak perspective Sebastian Thrun Perspective CS 223 B Computer Vision, Winter 2006 Marc Pollefeys 47
Summary: Perspective Laws 1. Perspective 2. Weak perspective 3. Orthographic Sebastian Thrun CS 223 B Computer Vision, Winter 2006 48
Limits for pinhole cameras Sebastian Thrun CS 223 B Computer Vision, Winter 2006 49
- Slides: 49