Input: Original intensity image. Target intensity image (i.e. a value sketch). Using Value Images to Adjust Intensity in 3D Renderings and Photographs.

Slides:



Advertisements
Similar presentations
A Model for Capturing NPR Shading from Art
Advertisements

Wang Tiles for Image and Texture Generation
Lecture 8 Transparency, Mirroring
Technische Universität München Computer Graphics SS 2014 Graphics Effects Rüdiger Westermann Lehrstuhl für Computer Graphik und Visualisierung.
Ray tracing. New Concepts The recursive ray tracing algorithm Generating eye rays Non Real-time rendering.
Contrast-Aware Halftoning Hua Li and David Mould April 22,
3D Graphics Rendering and Terrain Modeling
1 Computer Graphics Chapter 9 Rendering. [9]-2RM Rendering Three dimensional object rendering is the set of collective processes which make the object.
Topic 4 - Image Mapping - I DIGITAL IMAGING Course 3624 Department of Physics and Astronomy Professor Bob Warwick.
A Perceptual Heuristic for Shadow Computation in Photo-Realistic Images Wednesday, 2 August 2006 Peter VangorpOlivier DumontToon LenaertsPhilip Dutré.
 Engineering Graphics & Introductory Design 3D Graphics and Rendering REU Modeling Course – June 13 th 2014.
December 5, 2013Computer Vision Lecture 20: Hidden Markov Models/Depth 1 Stereo Vision Due to the limited resolution of images, increasing the baseline.
EE 7730 Image Segmentation.
Non-Photorealistic Rendering or More-Interesting Rendering (in a more positive sense) Presented by Adam Smith for CMPS160.
Curve Analogies Aaron Hertzmann Nuria Oliver Brain Curless Steven M. Seitz University of Washington Microsoft Research Thirteenth Eurographics.
Face Recognition Based on 3D Shape Estimation
A Novel 2D To 3D Image Technique Based On Object- Oriented Conversion.
Painterly Rendering for Animation Barbara J. Meier Walt Disney Feature Animation SIGGRAPH 96.
Computer graphics & visualization NPR – Non-photorealistic rendering.
Computer Graphics Shadows
Painterly Rendering for Animation – Barbara Meier
MIT Computer Science & Artificial Intelligence Laboratory The Choices We Make Frédo Durand MIT CSAIL.
Image Analogies Aaron Hertzmann (1,2) Charles E. Jacobs (2) Nuria Oliver (2) Brian Curless (3) David H. Salesin (2,3) 1 New York University 1 New York.
Multimedia Systems & Interfaces Karrie G. Karahalios Spring 2007.
UFCEKT-20-33D Modelling and Animation 3D Modelling & Animation Materials and Textures Maps.
Technology and Historical Overview. Introduction to 3d Computer Graphics  3D computer graphics is the science, study, and method of projecting a mathematical.
Computer Visualization BIM Curriculum 03. Topics  History  Computer Visualization Methods  Visualization Workflow  Technology Background.
Invitation to Computer Science 5th Edition
A Non-Photorealistic Model for Automatic Technical Illustration Amy Gooch Bruce Gooch Peter Shirley Elaine Cohen SIGGRAPH 1998 Presented by Anteneh.
Digital Face Replacement in Photographs CSC2530F Project Presentation By: Shahzad Malik January 28, 2003.
Light Using Texture Synthesis for Non-Photorealistic Shading from Paint Samples. Christopher D. Kulla, James D. Tucek, Reynold J. Bailey, Cindy M. Grimm.
Multiscale Moment-Based Painterly Rendering Diego Nehab and Luiz Velho
WIEN Stereo-based Image and Video Analysis for Multimedia Applications Application: “Computer-generated Stereoscopic Paintings“ M. Gelautz, E. Stavrakis,
Processing Images and Video for an Impressionist Effect Author: Peter Litwinowicz Presented by Jing Yi Jin.
December 4, 2014Computer Vision Lecture 22: Depth 1 Stereo Vision Comparing the similar triangles PMC l and p l LC l, we get: Similarly, for PNC r and.
Creating the Illusion of Motion in 2D Images. Reynold J. Bailey & Cindy M. Grimm Goal To manipulate a static 2D image to produce the illusion of motion.
03/05/03© 2003 University of Wisconsin Last Time Tone Reproduction If you don’t use perceptual info, some people call it contrast reduction.
Stylization and Abstraction of Photographs Doug Decarlo and Anthony Santella.
EECS 274 Computer Vision Segmentation by Clustering II.
CS654: Digital Image Analysis Lecture 18: Image Enhancement in Spatial Domain (Histogram)
Paint By Numbers The goal of a visual artist (Hagen): Without modeling detail, painters use brush strokes to: –Represent objects –Direct attention The.
Silicon Graphics, Inc. The Holodeck Interactive Ray Cache Greg Ward, Exponent Maryann Simmons, UCB.
Non-Photorealistic Rendering Motivation: Much of the graphical imagery created is not photographic in nature Particularly in some domains: –Art –Animation.
Fall 2009 Revised1 Texture Mapping (Intermediate) Jyun-Ming Chen.
Rendering Plant Leaves Faithfully Oliver Franzke (Dresden University of Technology) Oliver Deussen (University of Konstanz)
The Quotient Image: Class-based Recognition and Synthesis Under Varying Illumination T. Riklin-Raviv and A. Shashua Institute of Computer Science Hebrew.
Georgia Institute of Technology Initial Application of the Adaptive Grid Air Quality Model Dr. M. Talat Odman, Maudood N. Khan Georgia Institute of Technology.
AS-RIGID-AS-POSSIBLE SHAPE MANIPULATION
Extraction and remeshing of ellipsoidal representations from mesh data Patricio Simari Karan Singh.
- Laboratoire d'InfoRmatique en Image et Systèmes d'information
Orientable Textures for Image- Based Pen-And-Ink Illustration Michael P. Salisbury Michael T. Wong John F. Hughes David A. Salesin SIGGRAPH 1997 Andrea.
Yizhou Yu Texture-Mapping Real Scenes from Photographs Yizhou Yu Computer Science Division University of California at Berkeley Yizhou Yu Computer Science.
03/04/05© 2005 University of Wisconsin Last Time Tone Reproduction –Histogram method –LCIS and improved filter-based methods.
Nonphotorealistic rendering Computational Photography, Bill Freeman Fredo Durand May 9, 2006 Drawing from: NPR Siggraph 1999 course, Green et al.
Fine Tone Control in Hardware Hatching Matthew Webb Emil Praun Hugues Hoppe Adam Finkelstein Princeton University Microsoft Research Princeton University.
Non-Photorealistic Rendering FORMS. Model dependent Threshold dependent View dependent Outline form of the object Interior form of the object Boundary.
Eigen Texture Method : Appearance compression based method Surface Light Fields for 3D photography Presented by Youngihn Kho.
Color Image Segmentation Mentor : Dr. Rajeev Srivastava Students: Achit Kumar Ojha Aseem Kumar Akshay Tyagi.
Visible-Surface Detection Methods. To identify those parts of a scene that are visible from a chosen viewing position. Surfaces which are obscured by.
Processing Images and Video for An Impressionist Effect Automatic production of “painterly” animations from video clips. Extending existing algorithms.
(High Dynamic Range Imagery)
Week 14 - Wednesday CS361.
3D Graphics Rendering PPT By Ricardo Veguilla.
Fast Preprocessing for Robust Face Sketch Synthesis
Real-time Rendering Lab 4
Content-Sensitive Screening in Black and White
Image Based Modeling and Rendering (PI: Malik)
Real-time Rendering Shadow Maps
Real-time Volumetric Lighting in Participating Media
Artistic Rendering Final Project Initial Proposal
Presentation transcript:

Input: Original intensity image. Target intensity image (i.e. a value sketch). Using Value Images to Adjust Intensity in 3D Renderings and Photographs Reynold J. Bailey, Cindy M. Grimm Goal Technique for manipulating the intensity values in a 3D computer generated image or photograph. Motivation The value sketch is based on the lighting of the scene, but is not simply a faithful copy of it. Allows the artist to experiment with the placement of tone values. The use of value sketches leads to increased coherency and focus in the final image and also guides the viewer to the important elements of the image. Artists often make value sketches of a scene before beginning the final painting. Photographers and computer animators on the other hand, do not have the luxury of changing the tone values of a rendering to create the image they want. They must manipulate the lighting and/or the surface properties to achieve the desired effect. We present an alternative method for adjusting a 3D computer generated image or photograph based on a user provided value sketch. Related Work A tone mapping algorithm for high contrast images, Ashikhmin Interactive technical illustration, Gooch et al A user interface for interactive cinematic shadow design, Pellacini et al Local illumination environments for direct lighting acceleration, Fernandez et al Automatic lighting design using a perceptual quality metric, Shacked et al The lit sphere, Sloan et al Using texture synthesis for non- photorealistic shading from paint samples, Kulla et al Exposition Note: The value sketch for a given rendering can be created by using a 2D paint program or by using traditional art media. For the examples presented here, we avoid the problem of aligning the value image with the original image by using a 2D paint program to create a layer directly on top of the original image. Our system uses gradient descent to find a Value Mapping Function : such that: (1) pixels (x,y) is minimized. i.e. The absolute lighting in the resulting image corresponds to that of. (2) The relative lighting of the original image is preserved. i.e. is monotonically increasing. (3) Contrast in the resulting image is maximized. i.e. The slope of is maximized. A piece-wise linear function is used to represent. Pre-processing: The user selects the maximum number of desired divisions (N) for the Value Mapping Function. The system sorts the original pixels by increasing intensity values and uniformly assigns them to the N divisions of the function. Each division is given a unique ID. This is stored as an ID buffer. The system merges adjacent divisions if necessary to ensure that consecutive x- values of the function are strictly monotonically increasing (This prevents division by zero during slope calculations). During the optimization process, a weighted sum of three error functions (corresponding to the 3 desired properties of V) is evaluated. In general, it is not possible to match the value image exactly. One approach is to divide the image into regions that have similar value tones, and find a map that is ideal for each region. A color image segmentation technique proposed by Dorin Comaniciu and Peter Meer (1997) is used. Current / Future Work Other image segmentation techniques that allow for finer control are being explored. Exposition (cont’d) 3D Rendering Results OriginalValue sketch Segmented image (colored for illustration purposes) Note: For this example, the same weights were used for each segment. This causes the results for some segments to be closer to the target values than others. Evidence of this can be seen on the forehead where the lightest target area is not the lightest area in the result. We are currently trying to develop techniques that analyze the image and automatically generate weights for each segment. Result Photograph Results Original colorOriginal intensityValue sketchResult intensityResult color Value Mapping Function OriginalSegmented