Willem Mestdagh - Kevin Cordemans
Papers MS   Research PlayAnywhere: Compact   Interactive   Tabletop   Projection-Vision   System   [] Andy Wilson Thinsight: Integrated Optical Multi-touch Sensing through Thin Form-factor Displays  [] Bill Buxton,  Steve Hodges, ...
PlayAnywhere PlayAnywhere:  A Compact Interactive Tabletop Projection-Vision System  Andy Wilson at Microsoft Research, Redmond 2005
Introduction Looks great... But...  Calibration? Installation? Portability? Cost?
Mainstream Consumer Acceptance Multi-touch: new sensing and display technologies Ubiquitious computing  Portability Ease of installation Ease of use: any surface? Calibration Cost    Devise an easy-to-use, low-cost multi-touch system...
PlayAnywhere Front-projected computer vision-based interactive table system which uses a new projecting technology to obtain an exceptionnaly self-contained form factor Requirements: Quick set up: no calibration beyond factory No “installation” Usable on any flat surface Compact but still displaying and  sensing over a large surface (form factor) General purpose (Low cost)
Future? Canesta projection keyboard
PlayAnywhere What is it? Known problems in existing systems How does it work? PlayAnywhere configuration Projector Camera and IR Illuminant Image Processing Image Rectification Touch and Hover Analysis of shadow PlayAnywhere Visual Code Flow Move Computer Vision Pros and Cons
Known problems in existing systems
PlayAnywhere What is it? Known problems in existing systems How does it work? PlayAnywhere configuration Projector Camera and IR Illuminant Image Processing Image Rectification Touch and Hover Analysis of shadow PlayAnywhere Visual Code Flow Move Computer Vision Pros and Cons
PlayAnywhere Configuration All-in-One IR illuminant Projector (with mirrors) Camera with IR-passing-filter Prototype
Projector NEC WT600 DLP 40’’ diagonale afbeelding Voordelen van de plaats van de projector Gevaarlijke plaatsing projector is vermeden Wanneer camera en projector samen zijn geplaatst    geen nood om de camera te hercalibreren bij verplaatsing van het toestel Nadelen Beeldkwaliteit kan niet gegarandeerd worden,  want projectie-oppervlak is niet gekend
Camera en IR Illuminant De projector belicht de scene met zichtbaar straling De illuminant belicht de scene met infrarode straling Camera ontvangt alleen gereflecteerde infrarode stralen dankij de infrared-passing-filter die aan de camera bevestigd is (Camera blind voor reflecties van de projectorstraling)
How does it work? Image Processing Preparation: Image Rectification Computer Vision Techniques: Analysis of Shadow Hover and Touch Flow Move PlayAnywhere Visual Code
Image Rectification (1)    Remove distortion via standard bilinear interpolation techniques  Parameters for transformations need only be determined once (factory): Lens specs known Fixed topology device-plane (assumption) Setup  Good thing Bad thing Wide lens Maximize usable surface Creates barrel distortion Oblique position of camera Optimizes form factor (compact device) Produces projective distortion
Image Rectification (2)
Image Rectification (3) Input image and projected image are in a one-to-one correspondence, meaning: A rectangular  object on the surface appears as a rectangular object in the input image at the same (scaled) coordinates. Rectified image suited for computer vision techniques Limitation: objects further away from the unit will appear at a lower resolution (Minimum effective resolution is less than that of the acquired image)
Hover and Touch Detect touch without relying on special instrumentation of the surface, so that the device may operate on any available flat surface (goal) PlayAnywhere: exploite change in appearance of shadows as an object approaches the surface No tracking of fingers, ...
Hover and Touch thresholding operation Hover Candidate finger positions: highest on each of the distinct shadows which enter the scene from the bottom of the image Touch Analysis of shadow shapes: threshold width of finger shadow at a location slightly below the topmost point
Hover and Touch To limit false positives, candidate finger must lie on a shadow that extends to the bottom of the image    Objects on table can corrupt touch  detection by mimicing arm shadows,  meaning their shadows extends to the  bottom of the image Presently, only one finger per hand (more fingers requires more spohisticated analysis) Precision limited by image resolution
Flow Move Support direct manipulation of virtual objects  (doc’s, photo’s) Gestures that are natural to a specific activity E.g. resizing and rotating photo’s as if they were lying on a table PlayAnywhere: Optical flow techniques Based on local texture on the movig object No absolute position-based tracker, no widgets Instead simple statistics summarizing the flow field, computing only relative motion information
Optical Flow (1) Optical Flow computed from the most recently acquired rectified image I(t) and the previous image I(t-1) Mask images with grid Image (t-1) Image (t)
Optical Flow (2) For each (x,y) on a regular grid, the  integer  vector quantity  (dx, dy) is determined, such that:  image patch P centered on (x, y) at time t-1 most closely matches  image patch P’ centered on (x+dx,  y+dy) at time t. Image patches are compared by computing  pixelwise absolute differences  (low values    close match). If image patch P closely matches image patch P’, there is a good chance that the two patches are representing the same content, shifted to a different gridpoint in the interval.  (dx, dy) gives the direction of this shift. Image(t) (x+dx, y+dy) Image patch P’ Image(t-1) (x, y) Image patch P Resulting  vector
Optical Flow (3) For a given patch in the image, select (dx, dy) that minimizes Vector field can be used as an argument in a function to infer meaning from user’s gesture Resulting vector field
PlayAnywhere Visual Codes Ook mogelijk te werken met infrarood systemen in objecten    mogelijkheid tot werken met barcodes (derde vb met barcode)
Resultaat http://youtube.com/watch?v=YRfWhUnYyY8 Nu : Thingsight
Beschrijving “ Optical sensing system, fully integrated into a thin form factor display” Scherm met multi-touch systeem 2 mogelijkheden om multitouchsystemen uit te werken :  - een arrangement van electroden op het oppervlak - werken met cameras : voor of achter het scherm
Vergelijking Beide Touch-algoritme op basis van schaduwinval van handen Beide Handig te gebruiken en te verplaatsen Principe werking ongeveer gelijk    verschil : thinsight werkt met elektroden op het scherm, playanywhere werkt met camera’s boven het scherm Beiden hebben de mogelijkheid tot werken met externe objecten met een infrarode bron Voordeel bij thinsight : geen oppervlakte voor apparaat nodig

More Related Content

PPTX
3D scanner using kinect
PPTX
Concept of stereo vision based virtual touch
PDF
Stereo vision
PPTX
An Introduction to 3D TV
PDF
Viva3D Stereo Vision user manual en 2016-06
PPT
lecture4 raster details in computer graphics(Computer graphics tutorials)
PDF
Lecture1
PPT
ToolStone: Effective Use of The Physical Manipulation Vocabularies of Input D...
3D scanner using kinect
Concept of stereo vision based virtual touch
Stereo vision
An Introduction to 3D TV
Viva3D Stereo Vision user manual en 2016-06
lecture4 raster details in computer graphics(Computer graphics tutorials)
Lecture1
ToolStone: Effective Use of The Physical Manipulation Vocabularies of Input D...

What's hot (20)

PDF
Robots that see like humans
PPT
lecture1 introduction to computer graphics(Computer graphics tutorials)
PDF
Computer Graphics Notes
PPTX
Stereo vision
PDF
PR-228: Geonet: Unsupervised learning of dense depth, optical flow and camera...
PPTX
Hardware concept for graphics
PPTX
SIGGRAPH 2012 Computational Display Course - 3 Computational Light Field Disp...
PPTX
Datt 2501 week 11
PPTX
Introduction with graphics
PPTX
A concept of graphics
PPTX
PDF
Useful Tools for Making Video Games - Ogre (2008)
PDF
Compressive DIsplays: SID Keynote by Ramesh Raskar
PDF
291A_report_Hannah-Deepa-YunSuk
PPT
Animation
PPTX
HR3D: Content Adaptive Parallax Barriers
PPTX
3-D TECHNOLOGY
PPT
Animation
PPT
lecture2 computer graphics graphics hardware(Computer graphics tutorials)
PPTX
Datt 2500 week 11
Robots that see like humans
lecture1 introduction to computer graphics(Computer graphics tutorials)
Computer Graphics Notes
Stereo vision
PR-228: Geonet: Unsupervised learning of dense depth, optical flow and camera...
Hardware concept for graphics
SIGGRAPH 2012 Computational Display Course - 3 Computational Light Field Disp...
Datt 2501 week 11
Introduction with graphics
A concept of graphics
Useful Tools for Making Video Games - Ogre (2008)
Compressive DIsplays: SID Keynote by Ramesh Raskar
291A_report_Hannah-Deepa-YunSuk
Animation
HR3D: Content Adaptive Parallax Barriers
3-D TECHNOLOGY
Animation
lecture2 computer graphics graphics hardware(Computer graphics tutorials)
Datt 2500 week 11
Ad

Viewers also liked (7)

PDF
Extension
PPS
F Formiga
PPS
Holachicaguapa
PPT
Social and Collaborative Construction of Structured Knowledge WWW2007
PDF
Hans Leijström Disruptive Media
PPT
nuevas
PPT
poetry analysis
Extension
F Formiga
Holachicaguapa
Social and Collaborative Construction of Structured Knowledge WWW2007
Hans Leijström Disruptive Media
nuevas
poetry analysis
Ad

Similar to Presentatie Willem En Kevin (20)

PPT
Graphics display-devicesmod-1
PDF
Kinect v1+Processing workshot fabcafe_taipei
PDF
Keynote at Tracking Workshop during ISMAR 2014
PPTX
UNIT-6. Illumination model and surface renderingpptx
PPTX
Virtual_Reality_New.pptx
PPT
Introduction to Machine Vision
PDF
A Gesture Based Digital Art with Colour Coherence Vector Algorithm
PDF
K1802027780
PDF
10.1109@ecs.2015.7124874
DOC
Copy of 3 d report
PDF
CG2_HWcomputergraphicshardwareeeeeee.pdf
PPTX
Class[4][19th jun] [three js-camera&light]
PDF
Virtual Projection Interface
PPTX
computer Graphics
PPT
Pervasive Checkers
PPTX
Virtual Reality(full)
PPTX
Rendering Algorithms.pptx
PPT
Trytten computergraphics(1)
Graphics display-devicesmod-1
Kinect v1+Processing workshot fabcafe_taipei
Keynote at Tracking Workshop during ISMAR 2014
UNIT-6. Illumination model and surface renderingpptx
Virtual_Reality_New.pptx
Introduction to Machine Vision
A Gesture Based Digital Art with Colour Coherence Vector Algorithm
K1802027780
10.1109@ecs.2015.7124874
Copy of 3 d report
CG2_HWcomputergraphicshardwareeeeeee.pdf
Class[4][19th jun] [three js-camera&light]
Virtual Projection Interface
computer Graphics
Pervasive Checkers
Virtual Reality(full)
Rendering Algorithms.pptx
Trytten computergraphics(1)

Recently uploaded (20)

PPTX
Modernising the Digital Integration Hub
PDF
Abstractive summarization using multilingual text-to-text transfer transforme...
DOCX
search engine optimization ppt fir known well about this
PPTX
Benefits of Physical activity for teenagers.pptx
PDF
sustainability-14-14877-v2.pddhzftheheeeee
PDF
OpenACC and Open Hackathons Monthly Highlights July 2025
PDF
Consumable AI The What, Why & How for Small Teams.pdf
PDF
CloudStack 4.21: First Look Webinar slides
PPTX
Microsoft Excel 365/2024 Beginner's training
PDF
Produktkatalog für HOBO Datenlogger, Wetterstationen, Sensoren, Software und ...
PDF
How ambidextrous entrepreneurial leaders react to the artificial intelligence...
PDF
A review of recent deep learning applications in wood surface defect identifi...
PPTX
Chapter 5: Probability Theory and Statistics
PDF
Zenith AI: Advanced Artificial Intelligence
PPTX
MicrosoftCybserSecurityReferenceArchitecture-April-2025.pptx
PDF
A Late Bloomer's Guide to GenAI: Ethics, Bias, and Effective Prompting - Boha...
PPT
What is a Computer? Input Devices /output devices
PPTX
The various Industrial Revolutions .pptx
PDF
sbt 2.0: go big (Scala Days 2025 edition)
PPTX
Configure Apache Mutual Authentication
Modernising the Digital Integration Hub
Abstractive summarization using multilingual text-to-text transfer transforme...
search engine optimization ppt fir known well about this
Benefits of Physical activity for teenagers.pptx
sustainability-14-14877-v2.pddhzftheheeeee
OpenACC and Open Hackathons Monthly Highlights July 2025
Consumable AI The What, Why & How for Small Teams.pdf
CloudStack 4.21: First Look Webinar slides
Microsoft Excel 365/2024 Beginner's training
Produktkatalog für HOBO Datenlogger, Wetterstationen, Sensoren, Software und ...
How ambidextrous entrepreneurial leaders react to the artificial intelligence...
A review of recent deep learning applications in wood surface defect identifi...
Chapter 5: Probability Theory and Statistics
Zenith AI: Advanced Artificial Intelligence
MicrosoftCybserSecurityReferenceArchitecture-April-2025.pptx
A Late Bloomer's Guide to GenAI: Ethics, Bias, and Effective Prompting - Boha...
What is a Computer? Input Devices /output devices
The various Industrial Revolutions .pptx
sbt 2.0: go big (Scala Days 2025 edition)
Configure Apache Mutual Authentication

Presentatie Willem En Kevin

  • 1. Willem Mestdagh - Kevin Cordemans
  • 2. Papers MS Research PlayAnywhere: Compact Interactive Tabletop Projection-Vision System [] Andy Wilson Thinsight: Integrated Optical Multi-touch Sensing through Thin Form-factor Displays [] Bill Buxton, Steve Hodges, ...
  • 3. PlayAnywhere PlayAnywhere: A Compact Interactive Tabletop Projection-Vision System Andy Wilson at Microsoft Research, Redmond 2005
  • 4. Introduction Looks great... But... Calibration? Installation? Portability? Cost?
  • 5. Mainstream Consumer Acceptance Multi-touch: new sensing and display technologies Ubiquitious computing  Portability Ease of installation Ease of use: any surface? Calibration Cost  Devise an easy-to-use, low-cost multi-touch system...
  • 6. PlayAnywhere Front-projected computer vision-based interactive table system which uses a new projecting technology to obtain an exceptionnaly self-contained form factor Requirements: Quick set up: no calibration beyond factory No “installation” Usable on any flat surface Compact but still displaying and sensing over a large surface (form factor) General purpose (Low cost)
  • 8. PlayAnywhere What is it? Known problems in existing systems How does it work? PlayAnywhere configuration Projector Camera and IR Illuminant Image Processing Image Rectification Touch and Hover Analysis of shadow PlayAnywhere Visual Code Flow Move Computer Vision Pros and Cons
  • 9. Known problems in existing systems
  • 10. PlayAnywhere What is it? Known problems in existing systems How does it work? PlayAnywhere configuration Projector Camera and IR Illuminant Image Processing Image Rectification Touch and Hover Analysis of shadow PlayAnywhere Visual Code Flow Move Computer Vision Pros and Cons
  • 11. PlayAnywhere Configuration All-in-One IR illuminant Projector (with mirrors) Camera with IR-passing-filter Prototype
  • 12. Projector NEC WT600 DLP 40’’ diagonale afbeelding Voordelen van de plaats van de projector Gevaarlijke plaatsing projector is vermeden Wanneer camera en projector samen zijn geplaatst  geen nood om de camera te hercalibreren bij verplaatsing van het toestel Nadelen Beeldkwaliteit kan niet gegarandeerd worden, want projectie-oppervlak is niet gekend
  • 13. Camera en IR Illuminant De projector belicht de scene met zichtbaar straling De illuminant belicht de scene met infrarode straling Camera ontvangt alleen gereflecteerde infrarode stralen dankij de infrared-passing-filter die aan de camera bevestigd is (Camera blind voor reflecties van de projectorstraling)
  • 14. How does it work? Image Processing Preparation: Image Rectification Computer Vision Techniques: Analysis of Shadow Hover and Touch Flow Move PlayAnywhere Visual Code
  • 15. Image Rectification (1)  Remove distortion via standard bilinear interpolation techniques Parameters for transformations need only be determined once (factory): Lens specs known Fixed topology device-plane (assumption) Setup Good thing Bad thing Wide lens Maximize usable surface Creates barrel distortion Oblique position of camera Optimizes form factor (compact device) Produces projective distortion
  • 17. Image Rectification (3) Input image and projected image are in a one-to-one correspondence, meaning: A rectangular object on the surface appears as a rectangular object in the input image at the same (scaled) coordinates. Rectified image suited for computer vision techniques Limitation: objects further away from the unit will appear at a lower resolution (Minimum effective resolution is less than that of the acquired image)
  • 18. Hover and Touch Detect touch without relying on special instrumentation of the surface, so that the device may operate on any available flat surface (goal) PlayAnywhere: exploite change in appearance of shadows as an object approaches the surface No tracking of fingers, ...
  • 19. Hover and Touch thresholding operation Hover Candidate finger positions: highest on each of the distinct shadows which enter the scene from the bottom of the image Touch Analysis of shadow shapes: threshold width of finger shadow at a location slightly below the topmost point
  • 20. Hover and Touch To limit false positives, candidate finger must lie on a shadow that extends to the bottom of the image  Objects on table can corrupt touch detection by mimicing arm shadows, meaning their shadows extends to the bottom of the image Presently, only one finger per hand (more fingers requires more spohisticated analysis) Precision limited by image resolution
  • 21. Flow Move Support direct manipulation of virtual objects (doc’s, photo’s) Gestures that are natural to a specific activity E.g. resizing and rotating photo’s as if they were lying on a table PlayAnywhere: Optical flow techniques Based on local texture on the movig object No absolute position-based tracker, no widgets Instead simple statistics summarizing the flow field, computing only relative motion information
  • 22. Optical Flow (1) Optical Flow computed from the most recently acquired rectified image I(t) and the previous image I(t-1) Mask images with grid Image (t-1) Image (t)
  • 23. Optical Flow (2) For each (x,y) on a regular grid, the integer vector quantity (dx, dy) is determined, such that: image patch P centered on (x, y) at time t-1 most closely matches image patch P’ centered on (x+dx, y+dy) at time t. Image patches are compared by computing pixelwise absolute differences (low values  close match). If image patch P closely matches image patch P’, there is a good chance that the two patches are representing the same content, shifted to a different gridpoint in the interval. (dx, dy) gives the direction of this shift. Image(t) (x+dx, y+dy) Image patch P’ Image(t-1) (x, y) Image patch P Resulting vector
  • 24. Optical Flow (3) For a given patch in the image, select (dx, dy) that minimizes Vector field can be used as an argument in a function to infer meaning from user’s gesture Resulting vector field
  • 25. PlayAnywhere Visual Codes Ook mogelijk te werken met infrarood systemen in objecten  mogelijkheid tot werken met barcodes (derde vb met barcode)
  • 27. Beschrijving “ Optical sensing system, fully integrated into a thin form factor display” Scherm met multi-touch systeem 2 mogelijkheden om multitouchsystemen uit te werken : - een arrangement van electroden op het oppervlak - werken met cameras : voor of achter het scherm
  • 28. Vergelijking Beide Touch-algoritme op basis van schaduwinval van handen Beide Handig te gebruiken en te verplaatsen Principe werking ongeveer gelijk  verschil : thinsight werkt met elektroden op het scherm, playanywhere werkt met camera’s boven het scherm Beiden hebben de mogelijkheid tot werken met externe objecten met een infrarode bron Voordeel bij thinsight : geen oppervlakte voor apparaat nodig