feature-based image metamorphosis - computer sciencelazebnik/research/fall08/qi_mo.pdf · image...

32
Feature-Based Image Metamorphosis Thaddeus Beier Shawn Neely SIGGRAPH 1992

Upload: others

Post on 05-Oct-2020

9 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: Feature-Based Image Metamorphosis - Computer Sciencelazebnik/research/fall08/qi_mo.pdf · Image Morphing History Morphing is turning one image into another through a seamless transition

Feature-Based Image Metamorphosis

Thaddeus Beier Shawn NeelySIGGRAPH 1992

Page 2: Feature-Based Image Metamorphosis - Computer Sciencelazebnik/research/fall08/qi_mo.pdf · Image Morphing History Morphing is turning one image into another through a seamless transition

Image Morphing History

Morphing is turning one image into another through a seamless transition

Michael Jackson’s “Black or White”Cross-fading

Page 3: Feature-Based Image Metamorphosis - Computer Sciencelazebnik/research/fall08/qi_mo.pdf · Image Morphing History Morphing is turning one image into another through a seamless transition

morphing

cross-fadingImage morphing

image #1 image #2

warp warp

Page 4: Feature-Based Image Metamorphosis - Computer Sciencelazebnik/research/fall08/qi_mo.pdf · Image Morphing History Morphing is turning one image into another through a seamless transition

Image morphing

Morphing = warping + cross-dissolving

shape(geometric)

color(photometric)

Warp = feature specification + warp generation

Page 5: Feature-Based Image Metamorphosis - Computer Sciencelazebnik/research/fall08/qi_mo.pdf · Image Morphing History Morphing is turning one image into another through a seamless transition

Warp specification

• How can we specify the warp?3. Specify corresponding spline control points

• interpolate to a complete warping function

But we want to specify only a few points, not a grid

Page 6: Feature-Based Image Metamorphosis - Computer Sciencelazebnik/research/fall08/qi_mo.pdf · Image Morphing History Morphing is turning one image into another through a seamless transition

Warp specification

• How can we specify the warp?1. Specify corresponding points

• interpolate to a complete warping function• How do we do it?

Page 7: Feature-Based Image Metamorphosis - Computer Sciencelazebnik/research/fall08/qi_mo.pdf · Image Morphing History Morphing is turning one image into another through a seamless transition

Warp specification

• How can we specify the warp?2. Specify corresponding vectors

• interpolate to a complete warping function• The Beier & Neely Algorithm

Page 8: Feature-Based Image Metamorphosis - Computer Sciencelazebnik/research/fall08/qi_mo.pdf · Image Morphing History Morphing is turning one image into another through a seamless transition

Two basic styles

• Forward warping• Reverse mapping

Page 9: Feature-Based Image Metamorphosis - Computer Sciencelazebnik/research/fall08/qi_mo.pdf · Image Morphing History Morphing is turning one image into another through a seamless transition

Single line-pair PQ to P’Q’

Page 10: Feature-Based Image Metamorphosis - Computer Sciencelazebnik/research/fall08/qi_mo.pdf · Image Morphing History Morphing is turning one image into another through a seamless transition

Single Line-pair Examples

Page 11: Feature-Based Image Metamorphosis - Computer Sciencelazebnik/research/fall08/qi_mo.pdf · Image Morphing History Morphing is turning one image into another through a seamless transition

Multiple Lines

Length = length of the line segment, dist = distance to line segmenta, p, b – constants. What do they do?

Page 12: Feature-Based Image Metamorphosis - Computer Sciencelazebnik/research/fall08/qi_mo.pdf · Image Morphing History Morphing is turning one image into another through a seamless transition

Resulting warp (complex!)

Page 13: Feature-Based Image Metamorphosis - Computer Sciencelazebnik/research/fall08/qi_mo.pdf · Image Morphing History Morphing is turning one image into another through a seamless transition

Full Algorithm

Page 14: Feature-Based Image Metamorphosis - Computer Sciencelazebnik/research/fall08/qi_mo.pdf · Image Morphing History Morphing is turning one image into another through a seamless transition

Animation• Here's how you create an animated morph:• GenerateAnimation(Image0, L0[...],Image1, L1[...])• begin• foreach intermediate frame time t do• for i=1 to number of line-pairs do• L[i] = line t-th of the way from L0[i] to L1[i].• end• Warp0 = WarpImage( Image0, L0[...], L[...])• Warp1 = WarpImage( Image1, L1[...], L[...])• foreach pixel p in FinalImage do• FinalImage(p) = (1-t) Warp0(p) + t Warp1(p)• end• end• end

Page 15: Feature-Based Image Metamorphosis - Computer Sciencelazebnik/research/fall08/qi_mo.pdf · Image Morphing History Morphing is turning one image into another through a seamless transition

Interpolating Lines

• Method 1: interpolating endpoints• Method 2: interpolating midpoints, length

and orientation.

Page 16: Feature-Based Image Metamorphosis - Computer Sciencelazebnik/research/fall08/qi_mo.pdf · Image Morphing History Morphing is turning one image into another through a seamless transition

Results

Page 17: Feature-Based Image Metamorphosis - Computer Sciencelazebnik/research/fall08/qi_mo.pdf · Image Morphing History Morphing is turning one image into another through a seamless transition

Dynamic Scene

Page 18: Feature-Based Image Metamorphosis - Computer Sciencelazebnik/research/fall08/qi_mo.pdf · Image Morphing History Morphing is turning one image into another through a seamless transition

Algorithm summary

Page 19: Feature-Based Image Metamorphosis - Computer Sciencelazebnik/research/fall08/qi_mo.pdf · Image Morphing History Morphing is turning one image into another through a seamless transition

Morphing & matting• Extract foreground first to avoid artifacts in

the background

Page 20: Feature-Based Image Metamorphosis - Computer Sciencelazebnik/research/fall08/qi_mo.pdf · Image Morphing History Morphing is turning one image into another through a seamless transition

Uniform morphing

Page 21: Feature-Based Image Metamorphosis - Computer Sciencelazebnik/research/fall08/qi_mo.pdf · Image Morphing History Morphing is turning one image into another through a seamless transition

Non-uniform morphing

Page 22: Feature-Based Image Metamorphosis - Computer Sciencelazebnik/research/fall08/qi_mo.pdf · Image Morphing History Morphing is turning one image into another through a seamless transition

Procedural Transformation

Page 23: Feature-Based Image Metamorphosis - Computer Sciencelazebnik/research/fall08/qi_mo.pdf · Image Morphing History Morphing is turning one image into another through a seamless transition

Multi-source morphing

Page 24: Feature-Based Image Metamorphosis - Computer Sciencelazebnik/research/fall08/qi_mo.pdf · Image Morphing History Morphing is turning one image into another through a seamless transition

Manipulating Facial Appearance through Shape

and ColorDuncan A. Rowland and David I.

PerrettSt Andrews University

IEEE CG&A, September 1995

Page 25: Feature-Based Image Metamorphosis - Computer Sciencelazebnik/research/fall08/qi_mo.pdf · Image Morphing History Morphing is turning one image into another through a seamless transition

The Morphable Face Model

• shape vector S = (x1, y1, x2, …, yn)T

• appearance (texture) vector T = (R1, G1, B1, R2, …, Gn, Bn)T

Shape S

Appearance T

Page 26: Feature-Based Image Metamorphosis - Computer Sciencelazebnik/research/fall08/qi_mo.pdf · Image Morphing History Morphing is turning one image into another through a seamless transition

The Morphable face model• Assuming that we have m such vector pairs in full

correspondence, we can form new shapes Smodel and new appearances Tmodel as:

• If number of basis faces m is large enough to span the face subspace then:

• Any new face can be represented as a pair of vectors

∑=

=m

iiimodel a

1SS ∑

=

=m

iiimodel b

1

TT

Page 27: Feature-Based Image Metamorphosis - Computer Sciencelazebnik/research/fall08/qi_mo.pdf · Image Morphing History Morphing is turning one image into another through a seamless transition

Face averaging by morphing

average faces

http://www.beautycheck.de

Page 28: Feature-Based Image Metamorphosis - Computer Sciencelazebnik/research/fall08/qi_mo.pdf · Image Morphing History Morphing is turning one image into another through a seamless transition

Subpopulation means

• Examples:– Happy faces– Young faces– Asian faces– Etc.– Sunny days– Rainy days– Etc.– Etc.

Average male

Average female

Page 29: Feature-Based Image Metamorphosis - Computer Sciencelazebnik/research/fall08/qi_mo.pdf · Image Morphing History Morphing is turning one image into another through a seamless transition

The average face

Page 30: Feature-Based Image Metamorphosis - Computer Sciencelazebnik/research/fall08/qi_mo.pdf · Image Morphing History Morphing is turning one image into another through a seamless transition

Women In Arts

http://www.youtube.com/watch?v=nUDIoN-_Hxs

Page 31: Feature-Based Image Metamorphosis - Computer Sciencelazebnik/research/fall08/qi_mo.pdf · Image Morphing History Morphing is turning one image into another through a seamless transition

References• Thaddeus Beier, Shawn Neely, Feature-Based Image

Metamorphosis, SIGGRAPH 1992, pp35-42. • Detlef Ruprecht, Heinrich Muller, Image Warping with Scattered

Data Interpolation, IEEE Computer Graphics and Applications, March 1995, pp37-43.

• Seung-Yong Lee, Kyung-Yong Chwa, Sung Yong Shin, Image Metamorphosis Using Snakes and Free-Form Deformations, SIGGRAPH 1995.

• Seungyong Lee, Wolberg, G., Sung Yong Shin, Polymorph: morphing among multiple images, IEEE Computer Graphics and Applications, Vol. 18, No. 1, 1998, pp58-71.

• Peinsheng Gao, Thomas Sederberg, A work minimization approach to image morphing, The Visual Computer, 1998, pp390-400.

• George Wolberg, Image morphing: a survey, The Visual Computer, 1998, pp360-372.

Page 32: Feature-Based Image Metamorphosis - Computer Sciencelazebnik/research/fall08/qi_mo.pdf · Image Morphing History Morphing is turning one image into another through a seamless transition

Overview of Morphing Methods– Mesh Warping– Field Morphing– Radial Basis Function– Energy minimization– Multilevel Free-Form Deformation– Work minimization

Image Morphing: A Survey George Wolberg 1998