1 / 42

3D Photography

3D Photography. Acquisition of 3D images Registration between 3D images Creation of 3D models Texture-mapping 3D models Segmentation / classification of 3D models 3D object detection / localization Real-time applications Geometry Symmetry Detection …. Grading.

akrok
Download Presentation

3D Photography

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. 3D Photography • Acquisition of 3D images • Registration between 3D images • Creation of 3D models • Texture-mapping 3D models • Segmentation / classification of 3D models • 3D object detection / localization • Real-time applications • Geometry • Symmetry Detection • …

  2. Grading • 50 % for projects (homeworks + final group project) • 30% for presentations • 20% class participation

  3. 3D Photography complex / accurate representation of a 3D scene <<See slide from SIGGRAPH presentation>>

  4. Create geometric and photometric 3D models Use Range and Image Sensing Fusing image data Comprehensive system with automation Overview

  5. Computer Vision Sensors: Digital Cameras (2D) Range Scanners (3D) Illumination Vision System Physical 3D World Scene Description Geometry Reflectance

  6. 3D Photography & Graphics Scene Description Model of Illumination Geometry Reflectance Captured from Vision System Model of the Physical World • Modeling [Representation of 3D objects] • Rendering [Construction of 2D images from 3D models] • Animation [Simulating changes over time]

  7. 3D PHOTOGRAPHY EXAMPLE Ten scans were acquired of façade of Thomas Hunter Building (NYC) Automatic registration. Each scan has a different color. Registration details

  8. 3D PHOTOGRAPHY EXAMPLE 24 scans were acquired of façade of Shepard Hall (City College of NY)

  9. Data Acquisition • Leica ScanStation 2 • Spherical field of view. • Registered color camera. • Max Range: 300 meters. • Scanning time: 2-3 times faster • Accuracy: ~5mm per range point

  10. Data Acquisition, Leica Scan Station 2, Park Avenue and 70th Street, NY

  11. Applications • Virtual environment generation • Google Earth • acquire model for use in VRML, entertainment, etc • Realistic sets: movies and video games • Reverse engineering • acquiring a model from a part copying/modification • Part inspection • compare acquired model to “acceptable” model • 3D FAX • transmit acquired model to remote RP machine • Architectural site modeling • Urban Planning • Historical Preservation and Archaeology • Reverse Engineering of Buildings

  12. Data Acquisition Example Range Image. One million 3D points. Pseudocolor corresponds to laser intensity. Color Image of Thomas Hunter Building, New York City.

  13. Traditional Pipeline Range Images Brightness Images Segmentation 2D feature extraction 3D feature extraction Partial Model Complete Model Range-Intensity Registration Range-Range Registration FINAL MODEL FINAL PHOTOREALISTIC MODEL

  14. Other range sensors (some based on PrimeSense/Kinect) DotProduct Google Project Tango Prototype Matterport

  15. Course Format • Instructor will present recent topics and necessary background material • Each student will present one or two papers • Two projects • Final grade: 50% : projects 30% : presentation 20% : class participation

  16. Major Topics (tentantive) • Acquiring images: 2D and 3D sensors (digital cameras and laser range scanners). • 3D- and 2D-image registration. • Geometry: • representation of 3D models, • simplification of 3D models, • detection of symmetry. • Classification / Detection. • Rendering 3D models. • Image based rendering. • Texture mapping.

  17. Stereo Vision depth map

  18. Segmentation

  19. REGISTRATION pairwise & global

  20. 3D Modeling (Mesh or volumetric)

  21. Modelsimplification

  22. Passive techniques: Stereo and Structure from Motion

  23. 3D range to 2D image registration 2D image 3D scene

  24. 3D range to 2D image registration Corresponding 2D/proj. 3D lines Texture mapped 3D model

  25. TEXTURE MAP ANIMATION

  26. Symmetry Detection

  27. Image-Based Rendering • Chen and Williams (1993) - view interpolation • McMillan and Bishop (1995) - plenoptic modeling • Levoy and Hanrahan (1996) - light field rendering Slide by Ravi Ramamoorthi, Columbia University

  28. 3D Detection / Classification

  29. 3D Detection / Classification

  30. 3D Detection / Classification

  31. Dynamic Scenes Image sequence (CMU, Virtualized Reality Project) http://www.ri.cmu.edu/projects/project_144.html

  32. Dynamic Scenes Dynamic 3D model (CMU, Virtualized Reality Project) http://www.ri.cmu.edu/projects/project_144.html

  33. Dynamic Scenes Dynamic texture-mapped model (CMU, Virtualized Reality Project) http://www.ri.cmu.edu/projects/project_144.html

  34. Experimental Results • Details of merged segments Merged points Surface meshes

  35. Ball Pivoting Algorithm F. Bernardini, J. Mittleman, H. Rushmeier, C. Silva, G. Taubin. The ball-pivoting algorithm for surface reconstruction. IEEE Trans. on Vis. and Comp. Graph. 5 (4), pages 349-359, October-December 1999. A sequence of ball-pivoting operations. From left to right: A seed triangle is found; pivoting around an edge of the current front adds a new triangle to the mesh; after a number of pivoting operations, the activefront closes on itself; a final ball-pivoting completes the mesh. Closely related to alpha-shapes, Edelsbrunner 94

  36. Ball Pivoting Algorithm Results 3D mesh detail 3D mesh detail

  37. RECOVERED CAMERA POSITIONS

  38. GRAND CENTRAL (4)

  39. GRAND CENTRAL (5)

  40. 3D Modeling (people) Slide courtesy of Sebastian Thrun http://cs223b.stanford.edu Stanford Ioannis Stamos – CSCI 493.69 F08

  41. Head of Michelangelo’s David photograph 1.0 mm computer model

More Related