3D scanner using kinect
Download as PPTX, PDF2 likes2,473 views
This document discusses using the Kinect for 3D scanning. It begins with an introduction to 3D scanning and its importance. It then discusses the Kinect hardware and software tools used, including Point Cloud Library (PCL) and drivers. The implementation section explains the process of reading depth data, filtering, downsampling, plane removal, registration and outputting the final point cloud. Key steps include depth filtering, downsampling, plane removal, registration of multiple scans and using Geomagic software for refinement. The Kinect is used to capture 3D data which is then processed using PCL for applications such as animation, games and industrial use.
3D scanner using kinect
- 1. 3D Scanner Using kinect Mehdi Saman Booy Supervisor : Dr. Karimi
- 2. / 32 Contents • Introduction • 3D importance • Technologies • Kinect • PCL • Drivers • Implementation • Filtering • Registration 1
- 3. / 32 Introduction 2
- 4. / 32 3D Importance • 2D images Easily interpreted by human eyes Difficultly analyzing by computer 3
- 5. / 32 3D Importance (cont’d) • So, Third dimension is needed • We call it “depth” • Depth can be get from 3D scanner 4
- 6. / 32 3D Scanner Applications • Motion capture • Animations • Video games • Industrial • Reverse engineering • Quality check • Virtual reality • … 5
- 7. / 32 Technologies 6
- 8. / 32 Kinect • A webcam-style add-on peripheral (Wikipedia) • Enables users to control and interact with their console/computer without a controller • Originally for Xbox 360 • Also released for Windows and Xbox One • Kinetic + Connect = Kinect 7
- 9. Feb 2012 / 32 Kinect (history) 8 Oct 2011 Jun 2011 Feb 2011 Dec 2010 Nov 2010 Nov 2010
- 10. / 32 Kinect (history) 8 Kinect for Xbox 360 release Feb 2012 Oct 2011 Jun 2011 Feb 2011 Dec 2010 Nov 2010 Nov 2010
- 11. / 32 Kinect (history) Feb 2012 Oct 2011 Jun 2011 Feb 2011 Dec 2010 Nov 2010 Nov 2010 8 LibFreenect (Open source library) release
- 12. / 32 Kinect (history) 8 OpenNI (Open source driver) release Feb 2012 Oct 2011 Jun 2011 Feb 2011 Dec 2010 Nov 2010 Nov 2010
- 13. / 32 Kinect (history) 8 Announced to release Non-commercial Kinect SDK in spring 2011 Feb 2012 Oct 2011 Jun 2011 Feb 2011 Dec 2010 Nov 2010 Nov 2010
- 14. / 32 Kinect (history) 8 Non-commercial Kinect SDK release Feb 2012 Oct 2011 Jun 2011 Feb 2011 Dec 2010 Nov 2010 Nov 2010
- 15. Feb 2012 / 32 Kinect (history) 8 Oct 2011 Jun 2011 Feb 2011 Dec 2010 Nov 2010 Nov 2010 Announced to launching commercial version of Kinect (Kinect for Windows) with SDK
- 16. Feb 2012 / 32 Kinect (history) 8 Oct 2011 Jun 2011 Feb 2011 Dec 2010 Nov 2010 Nov 2010 Kinect for windows release
- 17. / 32 Kinect Hardware 9
- 18. / 32 Kinect Hardware 10
- 19. / 32 Kinect Hardware 10 Multi-array microphone
- 20. / 32 Kinect Hardware 10 IR laser emitter
- 21. / 32 Kinect Hardware 10 IR camera
- 22. Motorized tilt / 32 Kinect Hardware 10
- 23. USB cable / 32 Kinect Hardware 10
- 24. / 32 Kinect Hardware 10 RGB camera
- 25. / 32 PCL (Point Cloud Library) • Point Cloud: A set of data points in some coordinate system • PCL: A library to works with point clouds • Has a lots of functions for registration, filtering , … 11
- 26. / 32 Other Technologies • Geomagic Studio Good to works with 3D scanner outputs • OpenNI (Open Natural Interaction) Used to interact between Kinect and OS • Other drivers such as Microsoft official driver and Libfreenect 12
- 27. / 32 Use Kinect for Xbox 360 • As before said, Kinect is originally for Xbox 360 • Connect Kinect for Xbox 360 to PC is not easy • USB cable • Additional Driver • A lots of properties 13
- 28. / 32 Implementation 14
- 29. / 32 Our environment 15
- 30. / 32 Implementation Flowchart 16
- 31. / 32 Read data • Kinect depth image resolution : 640 x 480 • 307500 points (x, y, z for each point) • Shadow problem 17
- 32. / 32 Depth filtering • Kinect depth range : 50cm – 15m • Use passthrough filter • On z-axis between 1.5 , 15 18
- 33. / 32 Depth filtering Model after depth filtering between 1.5 and 15 m 19
- 34. / 32 Down sampling • Voxel: such as pixel in 2D image, each element of 3D • Use voxelGrid function • Change voxel size -> change down sample ratio 20
- 35. / 32 Down sampling Voxel grid leaf size is 0.5 cm 21
- 36. / 32 Plane removal • With RANSAC algorithm we can fit a mathematical model to set of points • Models : circle, sphere, plane, … • Fit a plane and delete that points 22
- 37. / 32 Cluster extraction • Still other objects or noises are exist • Points with distance lower than a threshold is a cluster • Delete all clusters except the biggest 23
- 38. / 32 Cluster extraction Clear object after filtering 24
- 39. / 32 Registration • Definition Aligning two or more images of the same scene One reference image Transform other images • The most important step 25
- 40. / 32 Registration (first try) • Angle of rotation is known • We can rotate object manually a) Original b) Expected c) happened 26 a b c
- 41. / 32 Registration (second try) 27 1 2 3 1) Translate object to center 2) Rotation 3) Translate object to its location
- 42. / 32 Registration (cont’d) • Another down sampling for each iteration is required • Number of iterations: n = 360 / θ • Note: Environment variables is so important. E.g. camera with center of object distance and angle of rotation 28
- 43. / 32 Point cloud result Result after registration process 29
- 44. / 32 Work with Geomagic • Better point cloud with Geomagic studio • Reduce noise • Refinement • Polygon mesh • Fill holes • Smoothness 30
- 45. / 32 Final result Result after work with Geomagic studio 31
- 46. / 32 Specially thanks to Dr. Karimi 32 Thank you
Editor's Notes
- #9: یک قطعه جانبی به سبک وب کم که اضافه میشه
- #19: See more detail at each part
- #20: See more detail at each part
- #21: See more detail at each part
- #22: See more detail at each part
- #23: See more detail at each part
- #24: See more detail at each part
- #25: See more detail at each part
- #32: Human eyes example
- #37: RANdom SAmple Consensus