Computer Vision CS302 Data Structures Dr. George Bebis

What is Computer Vision? Nice sunset! “Making computers see and understand”

Connections to other disciplines Computer Vision Image Processing Pattern Recognition Machine Learning Artificial Intelligence Robotics Psychology Neuroscience Computer Graphics

Image Processing

Image Processing (cont’d) Image enhancement Image Compression

Computer Graphics

Computer Graphics (cont’d) Image Output: Geometric Models Synthetic Camera projection, shading, lighting models

Computer Vision

Computer Vision (cont’d) Model Output: Real Scene CamerasImages

Why is Computer Vision Difficult? (1) It is a many-to-one mapping. –Inverse mapping has non-unique solution. –A lot of information is lost in the transformation from the 3D world to the 2D image. (2) It is computationally intensive. - A typical video is 30 frames / sec (3) We do not understand the recognition problem.

Viewpoint variations Michelangelo

Illumination changes

Scale changes

Deformations

Occlusions

Background clutter

Motion blurring

Intra-class variation

Local ambiguity

Applications Industry (visual inspection and assembly) Security and Surveillance (object detection, recognition, and tracking) Human Activity Recognition Traffic Monitoring and Analysis Robotics Medical Applications Many more …

Industrial Computer Vision (Machine Vision) Industrial computer vision systems work really well. Make strong assumptions about lighting conditions Make strong assumptions about the position of objects Make strong assumptions about the type of objects

Visual Inspection COGNEX

Optical character recognition (OCR) Digit recognition, AT&T labs Technology to convert scanned docs to text License plate readers Automatic check processing

Biometrics

Login without a password… Fingerprint scanners on many new laptops, other devices Face recognition systems begin to appear more widely

Face Recognition: Apple iPhoto

Face detection Many new digital cameras now detect faces –Canon, Sony, Fuji, …

Smile detection? Sony Cyber-shot® T70 Digital Still Camera

How the Afghan Girl was Identified by Her Iris Patterns Iris Biometrics

Hand-based Biometrics

Object Recognition (in supermarkets) LaneHawk by EvolutionRobotics “A smart camera is flush-mounted in the checkout lane, continuously watching for items. When an item is detected and recognized, the cashier verifies the quantity of items that were found under the basket, and continues to close the transaction. The item can remain under the basket, and with LaneHawk,you are assured to get paid for it… “

Mobile visual search: Google GogglesGoogle Goggles

Visual Surveillance and Human Activity Recognition Surveillance and security

Traffic Monitoring

Smart cars: –Vision systems currently in high-end BMW, GM, Volvo models. Mobileye

Automatic Panorama Stitching

Automatic Panorama Stitching (cont’d) find correspondences

3D Modeling

Medical Imaging Skin/Breast Cancer Detection 3D imaging MRI, CT Enable surgeons to visualize internal structures through an automated overlay of 3D reconstructions of internal anatomy on top of live video views of a patient. Image guided surgery Grimson et al., MIT

Robotics Semantic Robot Vision Challenge

Vision in space Vision systems (JPL) used for several tasks –Panorama stitching –3D terrain modeling –Obstacle detection, position tracking –For more, read “Computer Vision on Mars” by Matthies et al.Computer Vision on Mars NASA'S Mars Exploration Rover Spirit NASA'S Mars Exploration Rover Spirit captured this westward view from atop a low plateau where Spirit spent the closing months of 2007.

Vision-based Interaction and Games Nintendo Wii has camera-based IR tracking built in. See Lee’s work at CMU on clever tricks on using it to create a multi-touch display!Lee’s work at CMU multi-touch display Assistive technologies Kinect

Movie Special Effects Movie special effects Insert synthetic objects in real image sequences;. Change artificially the position or the orientation of a camera. Freeze a moving 3D scene.

Computer Vision Jobs !! Academia –MIT, UC-Berkeley, CMU, UIUC, USC …… UNR! National Labs and Government –Los Alamos National Lab –Lawrence Livermore National Lab –Navy, Air-force, Army Industry –Microsoft, Intel, IBM, Xerox, Compaq, Siemens, HP, TI, Motorola, Phillips, Honeywell, Ford See:

What skills would you need to succeed in this field? Strong programming skills (i.e., C, C++, Matlab) Very good knowledge of Data Structures and Algorithms Very good background in Mathematics, especially in: –Calculus –Linear Algebra –Probabilities and Statistics –Numerical Analysis –Geometry

Related Courses at UNR CS474/674 Image Processing and Interpretation (every Fall) CS485/685 Computer Vision (every Spring) CS486/686 Advanced Computer Vision (every Fall) CS479/679 Pattern Recognition (every other Spring) Special Topics –Biometrics, Object Recognition, Neural Networks, –Mathematical Methods for Computer Vision CS482/682 Artificial Intelligence CS773A Machine Intelligence CS791Q Machine Learning CS480/680 Computer Graphics

CS474/674 Image Processing 2 exams Homework 5-6 programming assignments noise eliminationlight correction contrast enhancement

CS485/685 Computer Vision Two exams Homework 5-6 programming assignments Paper presentations (grad students) face recognition3D reconstructionobject recognition

More information on Computer Vision Computer Vision Home Page UNR Computer Vision Laboratory