Download presentation
Presentation is loading. Please wait.
1
Nonphotorealistic rendering Computational Photography, 6.882 Bill Freeman Fredo Durand May 9, 2006 Drawing from: NPR Siggraph 1999 course, Green et al. npr_course_Sig99.pdf
2
Photorealism Physically realistic computer graphics rendering Images with photographic quality (eg Vermeer, 1632- 1675, accused by critics of being cold, inartistic, and displaying ‘spiritual poverty’). http://www.cs.utah.edu/npr/papers/npr_course_Sig99.pdf
3
Are these images non-photorealistic renderings? http://www.cs.utah.edu/npr/papers/npr_course_Sig99.pdf
4
Non-photorealistic rendering Expressive, artistic, painterly, interpretative rendering. Not aspiring to realism. Early work: natural media emulation –Pen and ink –Watercolor –Oil on canvas Attempts to capture the low-level style. Simulations of technical illustration. http://www.cs.utah.edu/npr/papers/npr_course_Sig99.pdf
5
NPAR 2002
6
Comparing photorealism and NPR (Stuart Green) http://www.cs.utah.edu/npr/papers/npr_course_Sig99.pdf
9
Statistical techniques to simulate expression http://www.cs.utah.edu/npr/papers/npr_course_Sig99.pdf
10
“Paintings are not solutions to well- posed problems…” http://www.cs.utah.edu/npr/papers/npr_course_Sig99.pdf
11
http://pages.cpsc.ucalgary.ca/~mario/npr/projects/sigg03/lec8/hand_1.pdf Daniel Teece
12
Organization of NPR methods Automated methods –2-d processing –3-d processing Interactive methods –2-d processing –3-d processing
13
Organization of NPR methods Automated methods –2-d processing –3-d processing Interactive methods –2-d processing –3-d processing
14
2/2.5 D, no user intervention http://www.cs.utah.edu/npr/papers/npr_course_Sig99.pdf
15
http://www.mrl.nyu.edu/publications/hertzmann-thesis/hertzmann-thesis-72dpi.pdf
16
Issues in image style translation Fitting Translation
17
http://people.csail.mit.edu/billf/papers/p33-t_freeman.pdf
19
Input traced line drawing This example will illustrate the tension between fitting and translation http://people.csail.mit.edu/billf/papers/p33-t_freeman.pdf
20
1-NN fit to input, style 1 Translation to style 2 Input drawing http://people.csail.mit.edu/billf/papers/p33-t_freeman.pdf
21
1-NN fit to input, style 1 Translation to style 2 Input drawing Bad fit, good translation http://people.csail.mit.edu/billf/papers/p33-t_freeman.pdf
22
5904-NN fit to input, style 1 Translation to style 2. Input drawing http://people.csail.mit.edu/billf/papers/p33-t_freeman.pdf
23
5904-NN fit to input, style 1 Translation to style 2. Input drawing Good fit, bad translation http://people.csail.mit.edu/billf/papers/p33-t_freeman.pdf
24
6-NN fit to input, style 1 Input drawing http://people.csail.mit.edu/billf/papers/p33-t_freeman.pdf
25
6-NN fit to input, style 1 Input drawing Translation to style 2 Good fit, good translation http://people.csail.mit.edu/billf/papers/p33-t_freeman.pdf
26
style 1 style 2style 3 http://people.csail.mit.edu/billf/papers/p33-t_freeman.pdf
28
6-NN fit to input, style 1 Translation to style 3 http://people.csail.mit.edu/billf/papers/p33-t_freeman.pdf
29
http://mrl.nyu.edu/projects/image-analogies/
30
http://mrl.nyu.edu/publications/image-analogies/analogies-72dpi.pdf
31
Image analogies applications
32
For painterly style translation, how get the A, A’ image pairs?
33
http://mrl.nyu.edu/projects/image-analogies/
36
Texture Transfer Take the texture from one object and “paint” it onto another object –This requires separating texture and shape –That’s HARD, but we can cheat –Assume we can capture shape by boundary and rough shading Then, just add another constraint when sampling: similarity to underlying image at that spot http://people.csail.mit.edu/billf/papers/efrosFreeman.pdf
37
Source texture Target image Source correspondence image Target correspondence image http://people.csail.mit.edu/billf/papers/efrosFreeman.pdf
39
A A’
41
B
42
B’
44
I think this one fails
45
Organization of NPR methods Automated methods –2-d processing –3-d processing Interactive methods –2-d processing –3-d processing
46
http://www.cs.utah.edu/npr/papers/npr_course_Sig99.pdf
49
Gooch and Gooch Concentrate on the material property and shading aspects of technical illustration.
50
Some characteristics of technical illustrations http://www.cs.utah.edu/npr/papers/npr_course_Sig99.pdf
52
Technical illustrations Lines
53
http://www.cs.utah.edu/npr/papers/npr_course_Sig99.pdf
54
Some parameterization dependent lines
55
http://www.cs.utah.edu/npr/papers/npr_course_Sig99.pdf
57
Line weight variations http://www.cs.utah.edu/npr/papers/npr_course_Sig99.pdf Equal weight Outer edges thicker Line weight varied to emphasize perspective
58
http://www.cs.utah.edu/npr/papers/npr_course_Sig99.pdf
61
Technical illustrations Shading
62
http://www.cs.utah.edu/npr/papers/npr_course_Sig99.pdf
66
Encoding surface orientation by color temperature http://www.cs.utah.edu/npr/papers/npr_course_Sig99.pdf
67
Direction dependent illumination color
68
Combining color-temp surface orientation coding with some tonal variations in object color http://www.cs.utah.edu/npr/papers/npr_course_Sig99.pdf
69
Parameter setting # 1 http://www.cs.utah.edu/npr/papers/npr_course_Sig99.pdf
70
Parameter setting # 2
71
http://www.cs.utah.edu/npr/papers/npr_course_Sig99.pdf
72
Metal object with anisotropic reflections http://www.cs.utah.edu/npr/papers/npr_course_Sig99.pdf
73
Metal object with anisotropic reflections http://www.cs.utah.edu/npr/papers/npr_course_Sig99.pdf “Lines are streaked in the direction of the axis of minimum curvature, parallel to the milling axis.”
74
http://www.cs.utah.edu/npr/papers/npr_course_Sig99.pdf
76
3D, little user intervention http://www.cs.utah.edu/npr/papers/npr_course_Sig99.pdf
Similar presentations
