Advancement of Graphics Hardware: 40 years since Sutherland s HMD

0
0
2533 days ago, 997 views
PowerPoint PPT Presentation
Advancement of Illustrations Equipment: 40 years since Sutherland's HMD. Henry FUCHS College of North Carolina at Church Slope Illustrations Equipment 2008 Sarajevo. Affirmation Help with arrangement of this discussion. Hideyuki Tamura, Standard Greg Welch, UNC Mark Mine, Disney

Presentation Transcript

Slide 1

Development of Graphics Hardware: a long time since Sutherland's HMD Henry FUCHS University of North Carolina at Chapel Hill Graphics Hardware 2008 Sarajevo

Slide 2

Acknowledgment Assistance with arrangement of this discussion Hideyuki Tamura, Canon Greg Welch, UNC Mark Mine, Disney Steve Molnar, NVIDIA Adam Lake, Intel

Slide 3

Evolution of Graphics H/W: from what point ? Ivan Sutherland's 1963 PhD Sketchpad Beginning of illustrations, however no new equipment Sutherland's 1965 "Extreme Display" thought A dream, yet no particular HW indicated Sutherland's 1968 3D HMD System Complete working framework, required all specific HW

Slide 4

Ivan E. Sutherland A Head-Mounted Three-Dimensional Display , 1968 Fall Joint Computer Conference Implemented every one of the segments of a constant intelligent 3D illustrations framework: Display gadget: stereo, optical transparent Image era: Real-time 3D point of view Head following: Real-time 6 DOF 3D show creation Hand-following & association: Real-time 3 DOF ('69 ?)

Slide 5

Procedure for Examining Evolution of Graphics HW For every subsystem: show, picture era, following.. 1968: what was actualized, best in class ~1988: test best in class 2008: current illustration subsystem Assessment: The fantasy in 1968/1970 Grade for where we are presently: A, B, C, D

Slide 6

Sutherland HMD: Overview

Slide 7

Images from Sutherland's '68 HMD

Slide 8

Display gadget on '68 HMD Optical transparent head-mounted show

Slide 9

HMD ca. 1985: VPL head-mount Closed show, not transparent LCD from SONY take TV wide edge optics (LEEP) (no contortion remedy)

Slide 10

HMD ca. 1995: video transparent HMD (UNC) Need for video, instead of optical, transparent

Slide 11

Current HM Displays shut, add up to VR Fakespace wide5 Video transparent: Canon/Mixed Reality Labs VH2002 InnerOptic (in light of UNC outline)

Slide 12

Assessment on HM Displays The fantasy at 1968/1970: Displays worked into our common eyeglasses; see virtual protests all through the eyeglass focal points (not a 'virtual inset') Grade for where we are presently: C critical advance off by a long shot to satisfactory

Slide 13

Image era: Sutherland's 1968 framework Real course of events drawing No shrouded line disposal Heroic work: everything worked out of entryway level chips 3D changes Clipping divider Graphics pipeline

Slide 14

Image era: ca 1988 Real-time full-shading raster picture era Rapid, reliable advance all through the 1990s 100s of polygons/sec in 1980 Million polygons/sec mid 1990s UNC Pixel-Planes 5 at siggraph 1991: 2M polygons/sec

Slide 15

Image era now Sophisticated shading impacts and a huge number of polygons/sec Commodity illustrations chips & loads up Rapid, proceeding with advance Image era processors getting to be distinctly universal –migrating even to cell phones

Slide 16

Image Generation Assessment Dream of 1968/1970: Realistic picture era Sutherland 1965 Ultimate show: can't recognize virtual and genuine items Grade for where we are presently: A+ Are we wrapped up? No! light up internationally; incorporate material science sim. with design; incorporate picture catch with picture show,.. Feels incredible for the field to do as such well.

Slide 17

Head following: Sutherland's 1968 framework Two following frameworks actualized ultrasound following with various emitters and beneficiaries mechanical tracker

Slide 18

Head following (ultrasonic) Sutherland's 1968 framework Ultrasonic following 3 transmitters on head-mount 4 recipients dangling from roof Measure stage changes Problem with vagueness of number of cycles of u/s signals Problem exacerbated by warming framework

Slide 19

Head following (mechanical) Sutherland's 1968 framework Vertical rotate in roof Universal joint on top Universal joint on base Shaft slides in and out "Sword of Damocles" Heavy and awkward Works

Slide 20

Head Tracking ~1988: Polhemus Magnetic Tracking Developed for head following in cockpit Very short proximity Not for strolling over a room Severe distorting of space with metal, with other attractive fields. Can be diminished with adjustment

Slide 21

Large region trackers: ~ 1991 UNC 10x12 ft space showed at Siggraph 1991 Multiple optical sensors on head, LEDs in roof tiles Lateral impact photograph diodes, significantly speedier than imagers One LED lit at once Predict most-helpful LED to light each vacancy

Slide 22

Head following Now Multiple great arrangements Optical Magnetic Inertial/optical blend Limited to instrumented region

Slide 23

Head following evaluation Dream of 1968/1970: Go anyplace, unhampered Grade for where we are presently: B Choice of arrangements available if willing to live with limitations Degraded execution if lose observable pathway contact (optical trackers) or close ferrous/metal items (attractive trackers) Can't go outside of instrumented zone, nor outside. Unfaltering advancement

Slide 24

Interaction/Hand following in Sutherland's HMD Mechanical outline with three reels of angling line mounted from roof All lines associated with top of (camera) hand-grasp Amount of line reeled out decides 3D area of hand-hold Problem of obstruction amongst head and hand trackers

Slide 25

Interaction/Hand following ca.1988 Polhemus attractive tracker (time-imparted to an objective for head-tracker) VPL DataGlove once in a while added to hand-tracker Severe twisting of followed space if client moves more than a couple steps

Slide 26

Interaction/Hand following now Multiple innovations accessible Accuracy & speed may not be satisfactory for specific applications UNC AR help for surgical needle direction Track ultrasound transducer Track needle

Slide 27

Assessment of Hand following Dream in 1968/1970: unhampered following of hand Grade for where we are currently: Incomplete Haven't had adequate number of genuine applications to judge - rest of framework hasn't been prepared utilizing Fred Brooks' denition of "What's genuine about Virtual Reality?" Someone other than creates pay to utilize the framework

Slide 28

3D display creation Sutherland's 1968 framework: inconsequentially basic

Slide 29

3D show creation ca. 1988 Manual model creation progressively costly/tedious with expanding capacity of picture era frameworks Feasible if 3D demonstrate should have been made as a major aspect of the application itself: CAD/CAM, chose therapeutic applications Automatic checking for chose objects: Cyberware 3D scanners

Slide 30

3D Model Creation Now Growing rundown of business and research abilities Computer vision: various pictures/video to 3D world models Laser scanners Motion catch frameworks

Slide 31

Assessment of Model Creation Dream in 1968/1970: Sutherland: Building model might be as tedious as building the genuine article Others: There must be a programmed path Grade for where we are currently C+ There are many apparatuses and frameworks, yet regularly still takes much dreary, manual exertion Compare the amount more work to digitizing this room and its substance than to rendering it

Slide 32

Assessment Summary Display: lastingly, next model will be better than average C Image era: unalloyed triumph! (be that as it may, don't release it to your head) A+ Head following: powerful however in restricted, arranged spaces B Interaction/Hand following: anticipating all the more genuine applications Incomplete Model era: frustratingly tedious C+

Slide 33

Opportunities: Look at bigger picture than a solitary module Head-mounted show configuration consolidated with picture era Fakespace Wide5: equipment geometric contortion amendment Mark Mine (Disney) actualizing more refined form in GPU Image era joined with head-following Model creation consolidated with picture era System issue spread over different modules: inactivity

Slide 34

Sutherland's vision of the HMD as the penultimate show still motivates

Slide 35

In conclusion: Don't compel your fantasies; there is so much cool illustrations HW to develop Stereo and autostereo shows Multi-projector/camera frameworks Vision catch for picture based rendering Haptics in the little and expansive Cell telephones w/design & vision: enchantment focal points in this present reality Telepresence frameworks (could do a whole hour on this) Finally– Alan Kay The most ideal approach to foresee what's to come is to concoct it THE END 35

SPONSORS