Forward interpolation on the GPU
|
|
- Angel Margery Green
- 5 years ago
- Views:
Transcription
1 Forward interpolation on the GPU GPU Computing Course 2010 Erik Ringaby
2 Questions Why is inverse interpolation not used here? Name at least one reason why forward interpolation is difficult to implement efficiently with CUDA? Why may my implementation run slower on <G80, <HD2000 models?
3 Agenda Background Main algorithm Forward vs. Inverse interpolation CUDA-implementation idea My implementation Coordinate calculation Results Conclusion
4 Computer Vision Laboratory Background Working on a project for rectifying rolling shutter images Original Corrected iphone 3GS
5 Global shutter With a global shutter, all lines have their integration time simultaneously A global reset is done just before readout Frame 1 readout Frame 2 integration Line 1 Line 2 Line 3 Line 4 Line n Readout Integration time
6 Rolling shutter All rows are exposed at different time, but the frame is stored as a single still image The readout resets the pixel content Frame 1 readout Frame 2 readout Line 1 Line 2 Line 3 Line 4 Line n Readout Integration time
7 Virtual global shutter
8 Computer Vision Laboratory Rectified RS-frame Original Corrected
9 Current implementation Most of the code written in Matlab KLT from OpenCV Forward interpolation in MEX Takes about 1-4s per frame
10 Main algorithm overview Find point correspondences between frames Estimate camera motion (rotation and/or translation) Transform rows/coordinates to one common coordinate system Interpolate image with new coordinates
11 Camera motion estimation Unfortunate the most time-consuming part is hard to parallelise Camera motion is estimated by minimising non-linear least squares Input is inter-frame point correspondences spread over whole frame (global) which should all be in the same cost function
12 Camera motion estimation Using Matlabs lsqnonlin as a black box and receives the motion parameters Big matrix with residuals Happy for parallelisation suggestions
13 Forward vs. Inverse interpolation When to use forward interpolation? Sometimes inverse mapping not available Here only an approximate inverse mapping exists
14 Forward vs. Inverse interpolation Neighbouring pixels within a row in the rectified image do not necessarily have the same transformation parameters Reconstruction comparison later
15 Forward interpolation At a first glance potentially easy to implement on a GPU due to a parallel nature Previous implemented via a MEX-function which is fed with the forward mapping
16 For each pixel in the input image, paint a blob in the output image The closest 3x3 grid are updated with:, 2 2 w( u) = exp( 0.5 u x ) / σ wb where CPU-implementation wr wg w u is grid location and x is sub-pixel location of pixel When all pixels have been written, divide with w to get RGB-image
17 CUDA implementation idea Local parallel problem, the input pixels does not need information from other pixels Even though CUDA supports scattered writes, gather operations tend to be much faster Since the output pixels don t know which input pixels that are going to contribute, it s difficult to write it as a gather operation
18 CUDA implementation idea Can be parallelised per row since we know how pixels are written within a row Different rows can however overlap and we have a race condition Atomic functions not supported on my GPU (G80) Even if it would support it, performance would decrease High-level function to determine problems
19 My implementation Geometry based What the GPU originally was designed for Move vertices with GLSL
20 First approach Set vertices at each rows end-points Send the forward mapping to the GPU and move the vertices to the correct position
21 Theoretical mistake Modification Not constant distance between pixels in the same row Need denser mesh for (row = 0; row <= height/rowres;row++){ for (col = 0; col <= width/colres; col++){ vertices[2*(nwidth+1)*row+2*col] = col*colres; vertices[2*(nwidth+1)*row+2*col+1] = row*rowres; } } texcoords[2*(nwidth+1)*row+2*col] = col*colres/width; texcoords[2*(nwidth+1)*row+2*col+1] = row*rowres/height; if ((col < nwidth) && (row < height/rowres)){ indices[4*nwidth*row+4*col] = (row * (nwidth+1)) + col; indices[4*nwidth*row+4*col+1] = (row * (nwidth+1)) + col + 1; indices[4*nwidth*row+4*col+2] = ((row+1) * (nwidth+1)) + col + 1; indices[4*nwidth*row+4*col+3] = ((row+1) * (nwidth+1)) + col; }
22 Final approach GPUs want to calculate data Instead of calculate coordinates on the CPU and sending them to the GPU, Calculate them directly on the GPU
23 How to calculate coordinates In the current main algorithm, rotations are assumed to be smooth across a frame From point correspondences the rotation for the bottom row is estimated Rotation is presented as a three element vector, n, with magnitude as rotation angle and direction as rotation axis 2 R = expm(n) = I + [ˆ] n x sinφ + [ˆ] n x(1 cosφ) r32 r23 n = logm( R) = φnˆ ~ / ~, ~, nˆ = n n n = r13 r31, φ = sin r21 r 12 1 ( n ~ / 2)
24 How to calculate coordinates R = expm(n) n = logm( R) I + [ˆ] n = φnˆ, sinφ + [ˆ] n 2 = x x (1 cosφ) r32 r23 ˆ ~ / ~, ~ n = n n n = r13 r31, φ = r21 r 12 sin 1 ( n ~ / 2) SLERP (Spherical Linear interpolation) Matrix exponential R n interp diff = expm( n = logm(expm( n 1 )expm( wn 1 diff ) )expm( n 2 w [ 0, 1] ))
25 How to calculate coordinates Each row gets its own rotation R through SLERP (w = currentrow / totalrows ) Transform all of them to a reference row with rotation R 0 (e.g. first or middle row) T 1 Forward mapping: x' = KR 0 R( x2) K x K intrinsic camera matrix Only need to send K and three values for last row rotation to the shader
26 GLSL GLSL is a perfect language for this Built-in data types like vec3, vec4 and mat3 Built-in functions length(), transpose(), cross(), matrix and vector multiplication
27 Forward vs. Inverse interpolation T Forward: x' = KR R( x Uses only vertex shader 0 2) K 1 x T 1 1 Inverse: x = K( R0 R( x2)) K x' Uses only fragment shader May be much slower on older cards (<G80, <HD2000) due to dedicated vertex/fragment processors instead of unified
28 Results Extreme case with ground-truth rotations to show differences Rolling shutter Forward Inverse No holes within mask
29 Results
30 Results Forward Mex ms Calculate coords in Matlab ~410ms Griddata usually ~19000ms (up to 1min) Inverse interpolation Matlab ~ ms Inverse interpolation GPU ~2ms Forward interpolation GPU ~15ms Just display image mesh ~15ms
31 Results The geometry is limiting the performance Downsample the grid Every 2 nd pixel per row 9-10ms Every 4 th pixel per row 5ms Every 8 th pixel per every 2 nd row 1-2ms, as fast as inverse interpolation
32 Results
33 GLSL still useful! Conclusion Forward interpolation on the GPU possible and fast (have been some discussion about this)
34 Questions?
35 Thank you!
PART IV: RS & the Kinect
Computer Vision on Rolling Shutter Cameras PART IV: RS & the Kinect Per-Erik Forssén, Erik Ringaby, Johan Hedborg Computer Vision Laboratory Dept. of Electrical Engineering Linköping University Tutorial
More information3D Geometry and Camera Calibration
3D Geometry and Camera Calibration 3D Coordinate Systems Right-handed vs. left-handed x x y z z y 2D Coordinate Systems 3D Geometry Basics y axis up vs. y axis down Origin at center vs. corner Will often
More informationEfficient Video Rectification and Stabilisation for Cell-Phones
Efficient Video Rectification and Stabilisation for Cell-Phones Erik Ringaby and Per-Erik Forssén Linköping University Post Print N.B.: When citing this work, cite the original article. The original publication
More informationPipeline Operations. CS 4620 Lecture Steve Marschner. Cornell CS4620 Spring 2018 Lecture 11
Pipeline Operations CS 4620 Lecture 11 1 Pipeline you are here APPLICATION COMMAND STREAM 3D transformations; shading VERTEX PROCESSING TRANSFORMED GEOMETRY conversion of primitives to pixels RASTERIZATION
More informationPipeline Operations. CS 4620 Lecture 14
Pipeline Operations CS 4620 Lecture 14 2014 Steve Marschner 1 Pipeline you are here APPLICATION COMMAND STREAM 3D transformations; shading VERTEX PROCESSING TRANSFORMED GEOMETRY conversion of primitives
More informationCS6670: Computer Vision
CS6670: Computer Vision Noah Snavely Lecture 7: Image Alignment and Panoramas What s inside your fridge? http://www.cs.washington.edu/education/courses/cse590ss/01wi/ Projection matrix intrinsics projection
More informationImage warping , , Computational Photography Fall 2017, Lecture 10
Image warping http://graphics.cs.cmu.edu/courses/15-463 15-463, 15-663, 15-862 Computational Photography Fall 2017, Lecture 10 Course announcements Second make-up lecture on Friday, October 6 th, noon-1:30
More information3.3 Interpolation of rotations represented by quaternions
3 S 3.3 Interpolation of rotations represented by quaternions : set of unit quaternions ; S :set of unit 3D vectors Interpolation will be performed on 3 S (direct linear interpolation produces nonlinear
More informationUnderstanding Shaders and WebGL. Chris Dalton & Olli Etuaho
Understanding Shaders and WebGL Chris Dalton & Olli Etuaho Agenda Introduction: Accessible shader development with WebGL Understanding WebGL shader execution: from JS to GPU Common shader bugs Accessible
More informationX. GPU Programming. Jacobs University Visualization and Computer Graphics Lab : Advanced Graphics - Chapter X 1
X. GPU Programming 320491: Advanced Graphics - Chapter X 1 X.1 GPU Architecture 320491: Advanced Graphics - Chapter X 2 GPU Graphics Processing Unit Parallelized SIMD Architecture 112 processing cores
More informationC P S C 314 S H A D E R S, O P E N G L, & J S RENDERING PIPELINE. Mikhail Bessmeltsev
C P S C 314 S H A D E R S, O P E N G L, & J S RENDERING PIPELINE UGRAD.CS.UBC.C A/~CS314 Mikhail Bessmeltsev 1 WHAT IS RENDERING? Generating image from a 3D scene 2 WHAT IS RENDERING? Generating image
More informationOPENGL RENDERING PIPELINE
CPSC 314 03 SHADERS, OPENGL, & JS UGRAD.CS.UBC.CA/~CS314 Textbook: Appendix A* (helpful, but different version of OpenGL) Alla Sheffer Sep 2016 OPENGL RENDERING PIPELINE 1 OPENGL RENDERING PIPELINE Javascript
More informationSUMMARY. CS380: Introduction to Computer Graphics Track-/Arc-ball Chapter 8. Min H. Kim KAIST School of Computing 18/04/06.
8/4/6 CS38: Introduction to Computer Graphics Track-/Arc-ball Chapter 8 Min H. Kim KAIST School of Computing Quaternion SUMMARY 2 8/4/6 Unit norm quats. == rotations Squared norm is sum of 4 squares. Any
More informationThe Rasterization Pipeline
Lecture 5: The Rasterization Pipeline Computer Graphics and Imaging UC Berkeley CS184/284A, Spring 2016 What We ve Covered So Far z x y z x y (0, 0) (w, h) Position objects and the camera in the world
More informationDynamic Geometry Processing
Dynamic Geometry Processing EG 2012 Tutorial Will Chang, Hao Li, Niloy Mitra, Mark Pauly, Michael Wand Tutorial: Dynamic Geometry Processing 1 Articulated Global Registration Introduction and Overview
More informationMore Mosaic Madness. CS194: Image Manipulation & Computational Photography. Steve Seitz and Rick Szeliski. Jeffrey Martin (jeffrey-martin.
More Mosaic Madness Jeffrey Martin (jeffrey-martin.com) CS194: Image Manipulation & Computational Photography with a lot of slides stolen from Alexei Efros, UC Berkeley, Fall 2018 Steve Seitz and Rick
More informationProgrammable Graphics Hardware
CSCI 480 Computer Graphics Lecture 14 Programmable Graphics Hardware [Ch. 9] March 2, 2011 Jernej Barbic University of Southern California OpenGL Extensions Shading Languages Vertex Program Fragment Program
More informationA Low Power, High Throughput, Fully Event-Based Stereo System: Supplementary Documentation
A Low Power, High Throughput, Fully Event-Based Stereo System: Supplementary Documentation Alexander Andreopoulos, Hirak J. Kashyap, Tapan K. Nayak, Arnon Amir, Myron D. Flickner IBM Research March 25,
More informationAdaptive Point Cloud Rendering
1 Adaptive Point Cloud Rendering Project Plan Final Group: May13-11 Christopher Jeffers Eric Jensen Joel Rausch Client: Siemens PLM Software Client Contact: Michael Carter Adviser: Simanta Mitra 4/29/13
More informationCamera Parameters and Calibration
Camera Parameters and Calibration Camera parameters U V W Ê Ë ˆ = Transformation representing intrinsic parameters Ê Ë ˆ Ê Ë ˆ Transformation representing extrinsic parameters Ê Ë ˆ X Y Z T Ê Ë ˆ From
More informationTSBK03 Screen-Space Ambient Occlusion
TSBK03 Screen-Space Ambient Occlusion Joakim Gebart, Jimmy Liikala December 15, 2013 Contents 1 Abstract 1 2 History 2 2.1 Crysis method..................................... 2 3 Chosen method 2 3.1 Algorithm
More informationTutorial (Beginner level): Orthomosaic and DEM Generation with Agisoft PhotoScan Pro 1.3 (without Ground Control Points)
Tutorial (Beginner level): Orthomosaic and DEM Generation with Agisoft PhotoScan Pro 1.3 (without Ground Control Points) Overview Agisoft PhotoScan Professional allows to generate georeferenced dense point
More informationobject (say a cube) will be made up of triangles, each with three vertices, each with known object coordinates.
hello world 3d: basic approach object (say a cube) will be made up of triangles, each with three vertices, each with known object coordinates. object coordinates of vertices will be put in an OpenGL buffer
More informationCamera Calibration. Schedule. Jesus J Caban. Note: You have until next Monday to let me know. ! Today:! Camera calibration
Camera Calibration Jesus J Caban Schedule! Today:! Camera calibration! Wednesday:! Lecture: Motion & Optical Flow! Monday:! Lecture: Medical Imaging! Final presentations:! Nov 29 th : W. Griffin! Dec 1
More informationN-Views (1) Homographies and Projection
CS 4495 Computer Vision N-Views (1) Homographies and Projection Aaron Bobick School of Interactive Computing Administrivia PS 2: Get SDD and Normalized Correlation working for a given windows size say
More informationAdvanced Graphics and Animation
Advanced Graphics and Animation Character Marco Gillies and Dan Jones Goldsmiths Aims and objectives By the end of the lecture you will be able to describe How 3D characters are animated Skeletal animation
More informationIntroduction to Computer Vision
Introduction to Computer Vision Michael J. Black Oct 2009 Motion estimation Goals Motion estimation Affine flow Optimization Large motions Why affine? Monday dense, smooth motion and regularization. Robust
More informationTutorial (Beginner level): Orthomosaic and DEM Generation with Agisoft PhotoScan Pro 1.3 (with Ground Control Points)
Tutorial (Beginner level): Orthomosaic and DEM Generation with Agisoft PhotoScan Pro 1.3 (with Ground Control Points) Overview Agisoft PhotoScan Professional allows to generate georeferenced dense point
More informationRigid ICP registration with Kinect
Rigid ICP registration with Kinect Students: Yoni Choukroun, Elie Semmel Advisor: Yonathan Aflalo 1 Overview.p.3 Development of the project..p.3 Papers p.4 Project algorithm..p.6 Result of the whole body.p.7
More information3D Geometry and Camera Calibration
3D Geometr and Camera Calibration 3D Coordinate Sstems Right-handed vs. left-handed 2D Coordinate Sstems ais up vs. ais down Origin at center vs. corner Will often write (u, v) for image coordinates v
More informationCSCI 5980: Assignment #3 Homography
Submission Assignment due: Feb 23 Individual assignment. Write-up submission format: a single PDF up to 3 pages (more than 3 page assignment will be automatically returned.). Code and data. Submission
More informationCS 4620 Midterm, October 23, 2018 SOLUTION
1. [20 points] Transformations CS 4620 Midterm, October 23, 2018 SOLUTION (a) Describe the action of each of the following matrices, as transformations in homogeneous coordinates, in terms of rotation,
More informationThe Illusion of Motion Making magic with textures in the vertex shader. Mario Palmero Lead Programmer at Tequila Works
The Illusion of Motion Making magic with textures in the vertex shader Mario Palmero Lead Programmer at Tequila Works Dark Ages before Textures in the Vertex Shader What is the Vertex Shader? A programmable
More informationSUMMARY. CS380: Introduction to Computer Graphics Texture Mapping Chapter 15. Min H. Kim KAIST School of Computing 18/05/03.
CS380: Introduction to Computer Graphics Texture Mapping Chapter 15 Min H. Kim KAIST School of Computing Materials SUMMARY 2 1 Light blob from PVC plastic Recall: Given any vector w (not necessarily of
More informationMultiview Stereo COSC450. Lecture 8
Multiview Stereo COSC450 Lecture 8 Stereo Vision So Far Stereo and epipolar geometry Fundamental matrix captures geometry 8-point algorithm Essential matrix with calibrated cameras 5-point algorithm Intersect
More informationCS GPU and GPGPU Programming Lecture 7: Shading and Compute APIs 1. Markus Hadwiger, KAUST
CS 380 - GPU and GPGPU Programming Lecture 7: Shading and Compute APIs 1 Markus Hadwiger, KAUST Reading Assignment #4 (until Feb. 23) Read (required): Programming Massively Parallel Processors book, Chapter
More informationHorn-Schunck and Lucas Kanade 1
Horn-Schunck and Lucas Kanade 1 Lecture 8 See Sections 4.2 and 4.3 in Reinhard Klette: Concise Computer Vision Springer-Verlag, London, 2014 1 See last slide for copyright information. 1 / 40 Where We
More informationCamera Model and Calibration
Camera Model and Calibration Lecture-10 Camera Calibration Determine extrinsic and intrinsic parameters of camera Extrinsic 3D location and orientation of camera Intrinsic Focal length The size of the
More informationCS4670: Computer Vision
CS4670: Computer Vision Noah Snavely Lecture 6: Feature matching and alignment Szeliski: Chapter 6.1 Reading Last time: Corners and blobs Scale-space blob detector: Example Feature descriptors We know
More informationCGT520 Lighting. Lighting. T-vertices. Normal vector. Color of an object can be specified 1) Explicitly as a color buffer
CGT520 Lighting Lighting Color of an object can be specified 1) Explicitly as a color buffer Bedrich Benes, Ph.D. Purdue University Department of Computer Graphics 2) Implicitly from the illumination model.
More informationUsing GPUs. Visualization of Complex Functions. September 26, 2012 Khaldoon Ghanem German Research School for Simulation Sciences
Visualization of Complex Functions Using GPUs September 26, 2012 Khaldoon Ghanem German Research School for Simulation Sciences Outline GPU in a Nutshell Fractals - A Simple Fragment Shader Domain Coloring
More informationGeometric Rectification of Remote Sensing Images
Geometric Rectification of Remote Sensing Images Airborne TerrestriaL Applications Sensor (ATLAS) Nine flight paths were recorded over the city of Providence. 1 True color ATLAS image (bands 4, 2, 1 in
More information12.2 Programmable Graphics Hardware
Fall 2018 CSCI 420: Computer Graphics 12.2 Programmable Graphics Hardware Kyle Morgenroth http://cs420.hao-li.com 1 Introduction Recent major advance in real time graphics is the programmable pipeline:
More informationSpeed up a Machine-Learning-based Image Super-Resolution Algorithm on GPGPU
Speed up a Machine-Learning-based Image Super-Resolution Algorithm on GPGPU Ke Ma 1, and Yao Song 2 1 Department of Computer Sciences 2 Department of Electrical and Computer Engineering University of Wisconsin-Madison
More informationAnimation Essentially a question of flipping between many still images, fast enough
33(70) Information Coding / Computer Graphics, ISY, LiTH Animation Essentially a question of flipping between many still images, fast enough 33(70) Animation as a topic Page flipping, double-buffering
More informationTSBK 07! Computer Graphics! Ingemar Ragnemalm, ISY
1(61) Information Coding / Computer Graphics, ISY, LiTH TSBK 07 Computer Graphics Ingemar Ragnemalm, ISY 1(61) Lecture 6 Texture mapping Skyboxes Environment mapping Bump mapping 2(61)2(61) Texture mapping
More informationCS179 GPU Programming Introduction to CUDA. Lecture originally by Luke Durant and Tamas Szalay
Introduction to CUDA Lecture originally by Luke Durant and Tamas Szalay Today CUDA - Why CUDA? - Overview of CUDA architecture - Dense matrix multiplication with CUDA 2 Shader GPGPU - Before current generation,
More informationAdvanced Lighting Techniques Due: Monday November 2 at 10pm
CMSC 23700 Autumn 2015 Introduction to Computer Graphics Project 3 October 20, 2015 Advanced Lighting Techniques Due: Monday November 2 at 10pm 1 Introduction This assignment is the third and final part
More informationShaders. Slide credit to Prof. Zwicker
Shaders Slide credit to Prof. Zwicker 2 Today Shader programming 3 Complete model Blinn model with several light sources i diffuse specular ambient How is this implemented on the graphics processor (GPU)?
More informationTransformation. Computer Graphics October. Dr Anton Gerdelan
Transformation Computer Graphics 4052 6 October Dr Anton Gerdelan gerdela@scss.tcd.ie Simplest Example Easiest way to scale our triangle? Easiest way to move our triangle? Demo time First Maths Revision
More informationReal-time Image-based Reconstruction of Pipes Using Omnidirectional Cameras
Real-time Image-based Reconstruction of Pipes Using Omnidirectional Cameras Dipl. Inf. Sandro Esquivel Prof. Dr.-Ing. Reinhard Koch Multimedia Information Processing Christian-Albrechts-University of Kiel
More informationReal-Time Shape Editing using Radial Basis Functions
Real-Time Shape Editing using Radial Basis Functions, Leif Kobbelt RWTH Aachen Boundary Constraint Modeling Prescribe irregular constraints Vertex positions Constrained energy minimization Optimal fairness
More informationMassively Parallel Non- Convex Optimization on the GPU Through the Graphics Pipeline
Massively Parallel Non- Convex Optimization on the GPU Through the Graphics Pipeline By Peter Cottle Department of Mechanical Engineering, 6195 Etcheverry Hall, University of California, Berkeley, CA 94720-1740,
More informationGraphics for VEs. Ruth Aylett
Graphics for VEs Ruth Aylett Overview VE Software Graphics for VEs The graphics pipeline Projections Lighting Shading VR software Two main types of software used: off-line authoring or modelling packages
More informationInverse Kinematics II and Motion Capture
Mathematical Foundations of Computer Graphics and Vision Inverse Kinematics II and Motion Capture Luca Ballan Institute of Visual Computing Comparison 0 1 A B 2 C 3 Fake exponential map Real exponential
More informationCS354 Computer Graphics Rotations and Quaternions
Slide Credit: Don Fussell CS354 Computer Graphics Rotations and Quaternions Qixing Huang April 4th 2018 Orientation Position and Orientation The position of an object can be represented as a translation
More informationMobile graphics API Overview
Mobile graphics API Overview Michael Doggett Department of Computer Science Lund University 2009 Michael Doggett and Tomas Akenine-Möller 1 Register Please check to see if your name is on the list, if
More informationCS 432 Interactive Computer Graphics
CS 432 Interactive Computer Graphics Lecture 4 3D Viewing Matt Burlick - Drexel University - CS 432 1 Reading Angel Chapters 3-4 Red Book Chapter 5, Appendix E Matt Burlick - Drexel University - CS 432
More informationComputer Vision I. Dense Stereo Correspondences. Anita Sellent 1/15/16
Computer Vision I Dense Stereo Correspondences Anita Sellent Stereo Two Cameras Overlapping field of view Known transformation between cameras From disparity compute depth [ Bradski, Kaehler: Learning
More informationSUMMARY. CS380: Introduction to Computer Graphics Projection Chapter 10. Min H. Kim KAIST School of Computing 18/04/12. Smooth Interpolation
CS38: Introduction to Computer Graphics Projection Chapter Min H. Kim KAIST School of Computing Smooth Interpolation SUMMARY 2 Cubic Bezier Spline To evaluate the function c(t) at any value of t, we perform
More informationThe Application Stage. The Game Loop, Resource Management and Renderer Design
1 The Application Stage The Game Loop, Resource Management and Renderer Design Application Stage Responsibilities 2 Set up the rendering pipeline Resource Management 3D meshes Textures etc. Prepare data
More informationGLSL Introduction. Fu-Chung Huang. Thanks for materials from many other people
GLSL Introduction Fu-Chung Huang Thanks for materials from many other people Shader Languages Currently 3 major shader languages Cg (Nvidia) HLSL (Microsoft) Derived from Cg GLSL (OpenGL) Main influences
More informationLinear Algebra and Image Processing: Additional Theory regarding Computer Graphics and Image Processing not covered by David C.
Linear Algebra and Image Processing: Additional Theor regarding Computer Graphics and Image Processing not covered b David C. La Dr. D.P. Huijsmans LIACS, Leiden Universit Februar 202 Differences in conventions
More information3D Transformations. CS 4620 Lecture 10. Cornell CS4620 Fall 2014 Lecture Steve Marschner (with previous instructors James/Bala)
3D Transformations CS 4620 Lecture 10 1 Translation 2 Scaling 3 Rotation about z axis 4 Rotation about x axis 5 Rotation about y axis 6 Properties of Matrices Translations: linear part is the identity
More informationInteractive Non-Linear Image Operations on Gigapixel Images
Interactive Non-Linear Image Operations on Gigapixel Images Markus Hadwiger, Ronell Sicat, Johanna Beyer King Abdullah University of Science and Technology Display-Aware Image Operations goal: perform
More informationIntroduction to Homogeneous coordinates
Last class we considered smooth translations and rotations of the camera coordinate system and the resulting motions of points in the image projection plane. These two transformations were expressed mathematically
More informationTexturing Theory. Overview. All it takes is for the rendered image to look right. -Jim Blinn 11/10/2018
References: Real-Time Rendering 3 rd Edition Chapter 6 Texturing Theory All it takes is for the rendered image to look right. -Jim Blinn Overview Introduction The Texturing Pipeline Example The Projector
More information2D rendering takes a photo of the 2D scene with a virtual camera that selects an axis aligned rectangle from the scene. The photograph is placed into
2D rendering takes a photo of the 2D scene with a virtual camera that selects an axis aligned rectangle from the scene. The photograph is placed into the viewport of the current application window. A pixel
More informationIntroduction to Computer Graphics. Image Processing (1) June 8, 2017 Kenshi Takayama
Introduction to Computer Graphics Image Processing (1) June 8, 2017 Kenshi Takayama Today s topics Edge-aware image processing Gradient-domain image processing 2 Image smoothing using Gaussian Filter Smoothness
More informationGame Programming. Bing-Yu Chen National Taiwan University
Game Programming Bing-Yu Chen National Taiwan University Character Motion Hierarchical Modeling Character Animation Motion Editing 1 Hierarchical Modeling Connected primitives 2 3D Example: A robot arm
More informationBroad field that includes low-level operations as well as complex high-level algorithms
Image processing About Broad field that includes low-level operations as well as complex high-level algorithms Low-level image processing Computer vision Computational photography Several procedures and
More informationEstimation of Automotive Pitch, Yaw, and Roll using Enhanced Phase Correlation on Multiple Far-field Windows
Estimation of Automotive Pitch, Yaw, and Roll using Enhanced Phase Correlation on Multiple Far-field Windows M. Barnada, C. Conrad, H. Bradler, M. Ochs and R. Mester Visual Sensorics and Information Processing
More informationGLSL Introduction. Fu-Chung Huang. Thanks for materials from many other people
GLSL Introduction Fu-Chung Huang Thanks for materials from many other people Programmable Shaders //per vertex inputs from main attribute aposition; attribute anormal; //outputs to frag. program varying
More informationBinocular Stereo Vision. System 6 Introduction Is there a Wedge in this 3D scene?
System 6 Introduction Is there a Wedge in this 3D scene? Binocular Stereo Vision Data a stereo pair of images! Given two 2D images of an object, how can we reconstruct 3D awareness of it? AV: 3D recognition
More information3D Transformations. CS 4620 Lecture Kavita Bala w/ prior instructor Steve Marschner. Cornell CS4620 Fall 2015 Lecture 11
3D Transformations CS 4620 Lecture 11 1 Announcements A2 due tomorrow Demos on Monday Please sign up for a slot Post on piazza 2 Translation 3 Scaling 4 Rotation about z axis 5 Rotation about x axis 6
More informationSIFT: SCALE INVARIANT FEATURE TRANSFORM SURF: SPEEDED UP ROBUST FEATURES BASHAR ALSADIK EOS DEPT. TOPMAP M13 3D GEOINFORMATION FROM IMAGES 2014
SIFT: SCALE INVARIANT FEATURE TRANSFORM SURF: SPEEDED UP ROBUST FEATURES BASHAR ALSADIK EOS DEPT. TOPMAP M13 3D GEOINFORMATION FROM IMAGES 2014 SIFT SIFT: Scale Invariant Feature Transform; transform image
More informationComputer Graphics - Treasure Hunter
Computer Graphics - Treasure Hunter CS 4830 Dr. Mihail September 16, 2015 1 Introduction In this assignment you will implement an old technique to simulate 3D scenes called billboarding, sometimes referred
More informationLecture 2. Shaders, GLSL and GPGPU
Lecture 2 Shaders, GLSL and GPGPU Is it interesting to do GPU computing with graphics APIs today? Lecture overview Why care about shaders for computing? Shaders for graphics GLSL Computing with shaders
More informationDX10, Batching, and Performance Considerations. Bryan Dudash NVIDIA Developer Technology
DX10, Batching, and Performance Considerations Bryan Dudash NVIDIA Developer Technology The Point of this talk The attempt to combine wisdom and power has only rarely been successful and then only for
More informationImage warping introduction
Image warping introduction 1997-2015 Josef Pelikán CGG MFF UK Praha pepca@cgg.mff.cuni.cz http://cgg.mff.cuni.cz/~pepca/ Warping 2015 Josef Pelikán, http://cgg.mff.cuni.cz/~pepca 1 / 22 Warping.. image
More informationLocally Weighted Least Squares Regression for Image Denoising, Reconstruction and Up-sampling
Locally Weighted Least Squares Regression for Image Denoising, Reconstruction and Up-sampling Moritz Baecher May 15, 29 1 Introduction Edge-preserving smoothing and super-resolution are classic and important
More informationShading and Illumination
Shading and Illumination OpenGL Shading Without Shading With Shading Physics Bidirectional Reflectance Distribution Function (BRDF) f r (ω i,ω ) = dl(ω ) L(ω i )cosθ i dω i = dl(ω ) L(ω i )( ω i n)dω
More informationZeyang Li Carnegie Mellon University
Zeyang Li Carnegie Mellon University Recap: Texture Mapping Programmable Graphics Pipeline Bump Mapping Displacement Mapping Environment Mapping GLSL Overview Perlin Noise GPGPU Map reflectance over a
More informationCS 4621 PPA3: Animation
CS 4621 PPA3: Animation out: Saturday 19 November 2011 due: Friday 2 December 2011 1 Introduction There are two parts to this assignment. In the first part, you will complete the implementation of key
More informationGraphics Hardware. Instructor Stephen J. Guy
Instructor Stephen J. Guy Overview What is a GPU Evolution of GPU GPU Design Modern Features Programmability! Programming Examples Overview What is a GPU Evolution of GPU GPU Design Modern Features Programmability!
More informationProgramming with OpenGL Part 3: Shaders. Ed Angel Professor of Emeritus of Computer Science University of New Mexico
Programming with OpenGL Part 3: Shaders Ed Angel Professor of Emeritus of Computer Science University of New Mexico 1 Objectives Simple Shaders - Vertex shader - Fragment shaders Programming shaders with
More informationHigh Performance Video Artifact Detection Enhanced with CUDA. Atul Ravindran Digimetrics
High Performance Video Artifact Detection Enhanced with CUDA Atul Ravindran Digimetrics Goals & Challenges Provide automated QC for digital video files with accuracy and minimum false positives Provide
More informationAn idea which can be used once is a trick. If it can be used more than once it becomes a method
An idea which can be used once is a trick. If it can be used more than once it becomes a method - George Polya and Gabor Szego University of Texas at Arlington Rigid Body Transformations & Generalized
More informationSpring 2011 Prof. Hyesoon Kim
Spring 2011 Prof. Hyesoon Kim Application Geometry Rasterizer CPU Each stage cane be also pipelined The slowest of the pipeline stage determines the rendering speed. Frames per second (fps) Executes on
More informationIMAGE-BASED RENDERING AND ANIMATION
DH2323 DGI17 INTRODUCTION TO COMPUTER GRAPHICS AND INTERACTION IMAGE-BASED RENDERING AND ANIMATION Christopher Peters CST, KTH Royal Institute of Technology, Sweden chpeters@kth.se http://kth.academia.edu/christopheredwardpeters
More informationOrthogonal Projection Matrices. Angel and Shreiner: Interactive Computer Graphics 7E Addison-Wesley 2015
Orthogonal Projection Matrices 1 Objectives Derive the projection matrices used for standard orthogonal projections Introduce oblique projections Introduce projection normalization 2 Normalization Rather
More informationSurfNet: Generating 3D shape surfaces using deep residual networks-supplementary Material
SurfNet: Generating 3D shape surfaces using deep residual networks-supplementary Material Ayan Sinha MIT Asim Unmesh IIT Kanpur Qixing Huang UT Austin Karthik Ramani Purdue sinhayan@mit.edu a.unmesh@gmail.com
More informationThe Graphics Pipeline
The Graphics Pipeline Ray Tracing: Why Slow? Basic ray tracing: 1 ray/pixel Ray Tracing: Why Slow? Basic ray tracing: 1 ray/pixel But you really want shadows, reflections, global illumination, antialiasing
More informationGPGPU, 1st Meeting Mordechai Butrashvily, CEO GASS
GPGPU, 1st Meeting Mordechai Butrashvily, CEO GASS Agenda Forming a GPGPU WG 1 st meeting Future meetings Activities Forming a GPGPU WG To raise needs and enhance information sharing A platform for knowledge
More information55:148 Digital Image Processing Chapter 11 3D Vision, Geometry
55:148 Digital Image Processing Chapter 11 3D Vision, Geometry Topics: Basics of projective geometry Points and hyperplanes in projective space Homography Estimating homography from point correspondence
More informationEE 584 MACHINE VISION
EE 584 MACHINE VISION Binary Images Analysis Geometrical & Topological Properties Connectedness Binary Algorithms Morphology Binary Images Binary (two-valued; black/white) images gives better efficiency
More informationDSP-Based Parallel Processing Model of Image Rotation
Available online at www.sciencedirect.com Procedia Engineering 5 (20) 2222 2228 Advanced in Control Engineering and Information Science DSP-Based Parallel Processing Model of Image Rotation ZHANG Shuang,2a,2b,
More informationInput Nodes. Surface Input. Surface Input Nodal Motion Nodal Displacement Instance Generator Light Flocking
Input Nodes Surface Input Nodal Motion Nodal Displacement Instance Generator Light Flocking The different Input nodes, where they can be found, what their outputs are. Surface Input When editing a surface,
More informationCS1114 Assignment 5, Part 1
CS4 Assignment 5, Part out: Friday, March 27, 2009. due: Friday, April 3, 2009, 5PM. This assignment covers three topics in two parts: interpolation and image transformations (Part ), and feature-based
More informationPhong Lighting & Materials. Some slides modified from: David Kabala Others from: Andries Van Damm, Brown Univ.
Phong Lighting & Materials Some slides modified from: David Kabala Others from: Andries Van Damm, Brown Univ. Lighting and Shading Lighting, or illumination, is the process of computing the intensity and
More information