Important concepts in binocular depth vision: Corresponding and non-corresponding points. Depth Perception 1. Depth Perception Part II

Similar documents
Binocular cues to depth PSY 310 Greg Francis. Lecture 21. Depth perception

Basic distinctions. Definitions. Epstein (1965) familiar size experiment. Distance, depth, and 3D shape cues. Distance, depth, and 3D shape cues

PERCEIVING DEPTH AND SIZE

Perception, Part 2 Gleitman et al. (2011), Chapter 5

USE R/G GLASSES. Binocular Combina0on of Images. What can happen when images are combined by the two eyes?

lecture 10 - depth from blur, binocular stereo

Think-Pair-Share. What visual or physiological cues help us to perceive 3D shape and depth?

Robert Collins CSE486, Penn State Lecture 08: Introduction to Stereo

Realtime 3D Computer Graphics Virtual Reality

Project 4 Results. Representation. Data. Learning. Zachary, Hung-I, Paul, Emanuel. SIFT and HoG are popular and successful.

Prof. Feng Liu. Spring /27/2014

Mahdi Amiri. May Sharif University of Technology

DEPTH PERCEPTION. Learning Objectives: 7/31/2018. Intro & Overview of DEPTH PERCEPTION** Speaker: Michael Patrick Coleman, COT, ABOC, & former CPOT

Multi-View Geometry (Ch7 New book. Ch 10/11 old book)

Processing Framework Proposed by Marr. Image

Stereo. Shadows: Occlusions: 3D (Depth) from 2D. Depth Cues. Viewing Stereo Stereograms Autostereograms Depth from Stereo

Stereovision. Binocular disparity

Virtual Reality ll. Visual Imaging in the Electronic Age. Donald P. Greenberg November 16, 2017 Lecture #22

Binocular Stereo Vision

Miniature faking. In close-up photo, the depth of field is limited.

The Absence of Depth Constancy in Contour Stereograms

Recap: Features and filters. Recap: Grouping & fitting. Now: Multiple views 10/29/2008. Epipolar geometry & stereo vision. Why multiple views?

Colorado School of Mines. Computer Vision. Professor William Hoff Dept of Electrical Engineering &Computer Science.

Pixels, Numbers, and Programs

Colorado School of Mines. Computer Vision. Professor William Hoff Dept of Electrical Engineering &Computer Science.

COMP 558 lecture 22 Dec. 1, 2010

Visual Secret Sharing Scheme with Autostereogram*

Subjects - Forty-nine students from Escola Superior de Tecnologia da Saúde de Lisboa

Stereo CSE 576. Ali Farhadi. Several slides from Larry Zitnick and Steve Seitz

CS 563 Advanced Topics in Computer Graphics Stereoscopy. by Sam Song

Stereoscopic Systems Part 1

Greatly enhanced visual detail and vividity. Accuracy based on mathematical derivation Disparity can function in isolation (RDS)

Stereo: Disparity and Matching

How to Shoot Good 3D with The Stereocam 3D Camcorder Adapter

Depth. Common Classification Tasks. Example: AlexNet. Another Example: Inception. Another Example: Inception. Depth

Vergence Micromovements and Depth Perception

Stereoscopic media. 3D is hot today. 3D has a long history. 3D has a long history. Digital Visual Effects Yung-Yu Chuang

Scaling of Rendered Stereoscopic Scenes

ABSTRACT Purpose. Methods. Results.

A Simple Viewfinder for Stereoscopic Video Capture Systems

P03 ray drawing with plane mirrors. November 06, Worlds Best Optical Illusion. Owls. Ray Diagrams.

Perception, 1999, volume 28, pages 143 ^ 153

Conflict? What conflict?

Seeing in three dimensions: the neurophysiology of stereopsis

The role of a local reference in stereoscopic. detection of depth relief

3D creativity for the Web

Stereospatial masking and aftereffect with normal and transformed random-dot patterns*

Robert Collins CSE486, Penn State. Robert Collins CSE486, Penn State. Image Point Y. O.Camps, PSU. Robert Collins CSE486, Penn State.

VS 117 LABORATORY III: FIXATION DISPARITY INTRODUCTION

Lecture 19. Depth Perception Reading Assignments: TMB2 7.1

Stereoscopic transparency: a test for binocular vision s disambiguating power 1

The Human Visual System!

The topics are listed below not exactly in the same order as they were presented in class but all relevant topics are on the list!

Last update: May 4, Vision. CMSC 421: Chapter 24. CMSC 421: Chapter 24 1

Stereo Graphics. Visual Rendering for VR. Passive stereoscopic projection. Active stereoscopic projection. Vergence-Accommodation Conflict

COMPUTER-GENERATED STEREOSCOPIC DISPLAYS

Visual areas in the brain. Image removed for copyright reasons.

Poor visibility of motion in depth is due to early motion averaging

2/26/2016. Chapter 23 Ray Optics. Chapter 23 Preview. Chapter 23 Preview

Announcements. Stereo

Physics Experiment 13

Are corresponding points fixed?

Vision: From Eye to Brain (Chap 3)

Mosaics wrapup & Stereo

Visual depth from motion parallax and eye pursuit

THE APERTURE PROBLEM IN STEREOPSIS

Hysteresis in human binocular fusion: temporalward and nasalward ranges

Do YOU See In 3D? DEFINITIONS

Introduction to visual computation and the primate visual system

Stereopsis based on monocular gaps: Metrical encoding of depth and slant without matching contours 1

Visual Rendering for VR. Stereo Graphics

Announcements. Stereo

EECS 442 Computer vision. Stereo systems. Stereo vision Rectification Correspondence problem Active stereo vision systems

Dispersion (23.5) Neil Alberding (SFU Physics) Physics 121: Optics, Electricity & Magnetism Spring / 17

3D Image Sensor based on Opto-Mechanical Filtering

CS6670: Computer Vision

CS5670: Computer Vision

Magic Pictures. 6pm Gresham College, 30 October, Harold Thimbleby Gresham Professor of Geometry

Computer Vision Lecture 17

Computer Vision Lecture 17

Stereo Optical Company Vision Tester Slide Package:

The efficiency of depth discrimination for non-transparent and transparent stereoscopic surfaces

Name Section Date. Experiment Reflection and Refraction

Human Perception of Objects

UNIT - V PERSPECTIVE PROJECTION OF SIMPLE SOLIDS

Durham E-Theses. Perceived Depth Control in Stereoscopic Cinematography SUN, GENG

1. What is the law of reflection?

Natural Viewing 3D Display

Stereoscopic matching and the aperture problem

Bio-inspired Binocular Disparity with Position-Shift Receptive Field

Chapter 26 Geometrical Optics

9. RAY OPTICS AND OPTICAL INSTRUMENTS

Chapter 34. Thin Lenses

Depth cue integration: stereopsis and image blur

Stereo Optical Company Vision Tester Slide Package:

Stevens High School AP Physics II Work for Not-school

Adaptive Image Sampling Based on the Human Visual System

Chapter 5. Projections and Rendering

4.5 Images Formed by the Refraction of Light

We are IntechOpen, the world s leading publisher of Open Access books Built by scientists, for scientists. International authors and editors

Transcription:

Depth Perception Part II Depth Perception 1 Binocular Cues to Depth Depth Information Oculomotor Visual Accomodation Convergence Binocular Monocular Static Cues Motion Parallax Perspective Size Interposition Shading Binocular Vision: Vision with Two Eyes Fixating an Object Binocular cues to depth: Binocular cues are based on the fact that we have two forward facing eyes that are laterally separated This provides slightly displaced images in each eye Fovea Fovea This information can be converted into a signal about relative depth Based on the geometry of the images reaching the eye Important concepts in binocular depth vision: Corresponding and non-corresponding points Fixation plane Horopter etinal disparity Diplopia Stereopsis Stereoacuity Stereopsis: Definitions Our brains convert overlapping flat images projected onto the retina of each eye into a 3-D model of the surrounding world. This creation of a 3-D world from the combining of information from the two eyes is called Stereopsis - from the Greek words stereos - for solid and opsis for vision - solid vision or solid sight. Stereopsis - is the ability to perceive depth or relative object distance based on retinal disparity.

Depth Perception 2 binocular stereopsis Not the most important cue for depth Why study it? Because its the only aspect of depth of which we have some physiological understanding Eavesdropping on binocular cells using electrophysiology The Early Stage of Stereopsis When we look at an object with two eyes we converge our eyes so the the image of the object falls on the fovea of each eye - the retinal locus with the highest resolving power. This convergence of the eyes so that the image of the object of interest falls on the foveas is called bifoveal fixation and is generally considered to be the first stage in binocular function. The foveas can be considered to be corresponding points on the two retinas. Thus any object you fixate will fall on corresponding points on the two retinas - i.e. the foveas. Corresponding and noncorresponding points When fixating, image of target falls on fovea of each eye The images of an object at the same distance as the fixation plane will fall on the same relative position in the two eyes Images that fall on different relative locations are said to fall on noncorresponding points Corresponding points and the horopter: The horopter is an imaginary plane through the fixation point that joins all corresponding points

The Horopter Points Falling on the Horopter Fall on Corresponding Points on the etinae Depth Perception 3 Non-corresponding Points and etinal Disparity If a target is closer or more distant than the fixation plane, its image falls on noncorresponding points in the two eyes etinal Disparity and Depth: There is a systematic relationship between the amount of retinal disparity on the retina and the distance of a target relative to the fixation plane If images fall on noncorresponding points, then there is retinal disparity and the potential for stereopsis definition of disparity = θ θ θ θ fixating here a b Fusion & Panum s Area The process by which we merge these retinally disparate images into a single percept is called fusion. Not all images that fall on disparate points lead to double vision-- which is also known as diplopia. ab a b left eye view right eye view There is a narrow region on either side of the horopter that includes all points in visual space that are fused into single images. This region is called Panum's area -the region where fusion occurs. perceived depth increases with increasing disparity (minus, closer than horopter, plus, farther than horopter)

Depth Perception 4 Panum s fusional space ocus of corresponding retinal points - Horopter Points that don't fall on the horopter fall on disparate (non corresponding) points in the two eyes. That is, objects located nearer or farther than the fixated target form images in different positions on each retina giving rise to disparity. The difference in the location of two retinal images of the same object is called binocular disparity. Crossed disparity Uncrossed disparity Stereoacuity andom Dot Stereogram The smallest disparity that can be resolved = Stereoacuity = 10-20 seconds of arc Invented by Bela Julesz 1956 emigre engineer from Hungary First innovative use of a computer for research in perception

How a andom Dot Stereogram Works: Depth Perception 5 random black and white pixels which are essentially the same in each eye some, however, are shifted laterally with respect to the others 1 0 1 0 1 0 1 1 0 1 0 1 1 0 0 1 1 1 1 1 1 0 1 1 0 0 1 1 0 1 left eye 1 0 1 0 1 0 1 0 1 0 1 1 1 0 1 1 1 1 1 1 0 1 0 1 0 0 1 1 0 1 right eye E E 1 0 1 0 1 0 1 0 1 0 1 0 row 1 1 1 0 1 0 1 1 0 1 0 1 1 row 2 Wheatstone s invention of the stereoscope (c. 1836) mirrors mirrors top view Some other methods to show stereo pictures Polaroid glasses method P+ P+ eft eye image mirrors ight eye image P- P- front view aluminized screen free fusion, w/o optical aids divergent convergent Each eye receives a separate image How does it work? notice that each eye receives a separate image of just two lines having a different separation. Brewster stereoscope

Plane of Fixation Disparity-Tuned Cell esponses C A B C' A' B' Depth Perception 6 Subject fixating B Individual neurons were tuned for different amounts and directions of disparity Stimulation ocation (A,A') (B,B') (C,C') Stimulation ocation Cell esponses "Near" cell Cell tuned to fixation plane "Far" cell Several different classes of neuron, some finely tuned for small amounts of disparity, others simply responding to near or far Autostereograms: In autostereograms we use our vergence eye movements as the stereoscope A Simplified Example of How An Autostereogram Works By converging or diverging we shift the image in one eye relative to the other With the correct amount of vergence we are now superimposing two parts of a repeating image which has been designed to contain disparity when viewed this way Simplified Magic Eye Autostereogram

Depth Perception 7 Binocular parallax a b Notice the difference in angular separation Size constancy Size constancy The perception of size is closely related to the perception of distance. The brain is remarkably good at compensating for changes in retinal image size with distance in order to keep the perceived size constant Why does someone walking away not appear to shrink?

Size constancy: Given the size of the image on the retina (visual angle) and its distance, it is possible to compute the physical size of an object Depth Perception 8 The Holway-Boring experiment: Size constancy is the mechanism that makes this computation Holway & Boring demonstrated the crucial importance of depth perception in an experiment Observer views Test Disks located at different distances Task is to adjust size of Comparison Disk to match physical size of Test Disk Test disks all set to subtend 1 o of visual angle Test under several condition in which the availability of depth cues is varied Observers matched closer to visual angle as cues removed elationship between size perception and perceived distance: Emmert s aw Generate afterimage on retina View afterimage against surfaces at different distances Note changed size of afterimage The perceived size of an afterimage is related to the distance of the viewing surface from the eye

Depth Perception 9 Emmer t's law: Sp = k x Sr x Dp (Sp = pe rceived size; Sr = retinal size; Dp = perceived d istance; k = constant) Ponzo Illusion This picture looks odd because the size and distance cues are in conflict Size perception and visual illusions: A number of visual illusions may result from the misapplication of constancy scaling Gregory has argued that the misjudgement of size is because the illusory figure contains information that activates the constancy scaling mechanism Consequently, an object is seen as larger or smaller than it should be Muller-yer Illusion: If the arrowheads are seen as internal and external contours, the closer, external corner should appear bigger

Depth Perception 10 The moon illusion: Moon (or sun) seems larger at the horizon than at the zenith ecognised in classical times, many theories Assumption: if two objects have the same retinal image size, the one that appears closer will look smaller That means horizon moon must look more distant Some evidence that horizon looks further away than zenith sky Current most-accepted explanation in terms of apparent distance, although issue is still controversial If zenith sky appears closer, then moon will seem to be smaller