vision recognition

26
Vision Recognition Vision Recognition Neil Agarwal Neil Agarwal Bailey Zhang Bailey Zhang Zack Sorenson Zack Sorenson Ben Schneider Ben Schneider

Upload: honorato-andrews

Post on 30-Dec-2015

37 views

Category:

Documents


0 download

DESCRIPTION

Vision Recognition. Neil Agarwal Bailey Zhang Zack Sorenson Ben Schneider. What is Computer Vision?. Definition Computational Vision Attempts to understand visual perception and understanding Machine Vision Attempts to create practical application for computer vision. - PowerPoint PPT Presentation

TRANSCRIPT

Vision RecognitionVision RecognitionNeil AgarwalNeil AgarwalBailey ZhangBailey Zhang

Zack SorensonZack SorensonBen SchneiderBen Schneider

What is Computer Vision?What is Computer Vision?

DefinitionDefinition Computational VisionComputational Vision

Attempts to understand visual perception and Attempts to understand visual perception and understandingunderstanding

Machine VisionMachine Vision Attempts to create practical application for Attempts to create practical application for

computer visioncomputer vision

Vision Recognition & AIVision Recognition & AI

Creating useful autonomous agentsCreating useful autonomous agents Human/computer interfacingHuman/computer interfacing

Gesture recognitionGesture recognition Handwriting recognitionHandwriting recognition Passive observation & analysisPassive observation & analysis Understanding biological visionUnderstanding biological vision

Honda’s Honda’s ASIMOASIMO

GRACE the RobotGRACE the Robot

Converting handwritten figures into figures usable by a computer

Current ImplementationCurrent ImplementationOptical Character Recognition (OCR)Optical Character Recognition (OCR)

Types of OCRTypes of OCRCharacter-based word recognition Character-based word recognition Segmentation-based word recognition Segmentation-based word recognition Word-shape recognition Word-shape recognition

Current Limitations of OCRCurrent Limitations of OCR Underlines in text Underlines in text ‘ ‘I’I’ and and ‘i’‘i’ confusions confusions Distortion of Distortion of ‘e’‘e’

Examples of OCR softwareExamples of OCR software Calera Wordscan Plus 1.0Calera Wordscan Plus 1.0

Caere Omnipage Professional 3.0Caere Omnipage Professional 3.0 Xerox Imaging Systems AccuText 3.0Xerox Imaging Systems AccuText 3.0

Current Implementation (cont.)Current Implementation (cont.)

““ScarecrowScarecrow”” Floating, autonomous robot platform powered Floating, autonomous robot platform powered

by solar panels by solar panels Searches for birds using its vision systemSearches for birds using its vision system Goes around, when its senses birds, bumps Goes around, when its senses birds, bumps

them or shoots water at themthem or shoots water at them Floats in catfish ponds to protect the stock Floats in catfish ponds to protect the stock

which is frequently depleted by hungry birds which is frequently depleted by hungry birds

Successes in Successes in Computer/MachineComputer/Machine

VisionVision Machine Vision, the name for Computer Machine Vision, the name for Computer

Vision that is applied specifically to robots Vision that is applied specifically to robots or machines, is extremely relevant in or machines, is extremely relevant in industry.industry.

Most Computer Vision technology that Most Computer Vision technology that affects you is in fact Machine Vision.affects you is in fact Machine Vision.

Machine Vision and YouMachine Vision and You

Machine Vision is being developed to control the quality Machine Vision is being developed to control the quality of products, detecting defects automaticallyof products, detecting defects automatically

The technology is reaching a point where people can be The technology is reaching a point where people can be tracked in 3D, leading to more robust security systems.tracked in 3D, leading to more robust security systems.

And for your entertainment, Machine vision is being And for your entertainment, Machine vision is being developed to track actual movements, for more realistic developed to track actual movements, for more realistic and exciting gamesand exciting games

Other Machine Vision FieldsOther Machine Vision Fields

Agriculture, Architecture, Character Agriculture, Architecture, Character Recognition, Cultural, Forensic, Medical, Recognition, Cultural, Forensic, Medical, Image Processing, Reverse Engineering, Image Processing, Reverse Engineering, Military, Navigation, Remote Sensing, Military, Navigation, Remote Sensing, Safety, SportsSafety, Sports

Successes in Computer VisionSuccesses in Computer Vision

Computer Vision is broader than Computer Vision is broader than Machine Vision.Machine Vision.

There are many successes in Computer There are many successes in Computer Vision; many, many, many successesVision; many, many, many successes

Development/Successes in Development/Successes in Computer Vision TechnologyComputer Vision Technology

SUCCESSORSUCCESSOR

General SuccessesGeneral Successes

Marr’s Theory: shape info from images.Marr’s Theory: shape info from images.

Marr’s TheoryMarr’s Theory

The geometry The geometry The reflectance of the visible surfacesThe reflectance of the visible surfaces The illuminationThe illumination The viewpointThe viewpoint

More MarrMore Marr

Problems & SetbacksProblems & Setbacks

Finding 3D objects in 2D imagesFinding 3D objects in 2D images Making raster images usable for a computerMaking raster images usable for a computer

Incomplete understanding of human visionIncomplete understanding of human vision

What’s Currently Being DoneWhat’s Currently Being Done

Penn State Computer Vision LabPenn State Computer Vision Lab Berkeley GroupBerkeley Group

Penn StatePenn State

Hand gesture Hand gesture recognition as recognition as supplement to speech supplement to speech recognition.recognition.

Text recognition and Text recognition and clarification in video.clarification in video.

Text ClarificationText Clarification

Possible example need for technologyPossible example need for technology

BerkeleyBerkeley

Automatic detection of Automatic detection of human nudeshuman nudes

Image content Image content recognitionrecognition

Image Content AnalysisImage Content Analysis

Future of Vision RecognitionFuture of Vision Recognition

PC’s Recognizing Users (Microsoft Project)PC’s Recognizing Users (Microsoft Project) When a user sits in front of the computer, it will When a user sits in front of the computer, it will

bring up his work, etc.bring up his work, etc. PC equipped with a camera on topPC equipped with a camera on top Long term: aiming for a system that can Long term: aiming for a system that can

recognize eye movements, so that software recognize eye movements, so that software could be controlled by eye movement could be controlled by eye movement

Only a couple years from reaching mass Only a couple years from reaching mass market?market?

Future of Vision Recognition (cont.)Future of Vision Recognition (cont.)

Augmented RealityAugmented Reality

Real world

Augmented world

Any Questions?Any Questions?