FeatureBased Image Metamorphosis Thaddeus Beier Shawn Neely SIGGRAPH
Feature-Based Image Metamorphosis Thaddeus Beier Shawn Neely SIGGRAPH 1992
Image Morphing History Morphing is turning one image into another through a seamless transition Michael Jackson’s “Black or White” Cross-fading
Image morphing image #1 warp cross-fading morphing image #2 warp
Image morphing Morphing = warping + cross-dissolving color (photometric ) Warp = feature specification + warp generation shape (geometric)
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
Warp specification • How can we specify the warp? 1. Specify corresponding points • interpolate to a complete warping function • How do we do it?
Warp specification • How can we specify the warp? 2. Specify corresponding vectors • interpolate to a complete warping function • The Beier & Neely Algorithm
Two basic styles • Forward warping • Reverse mapping
Single line-pair PQ to P’Q’
Single Line-pair Examples
Multiple Lines Length = length of the line segment, dist = distance to line segment a, p, b – constants. What do they do?
Resulting warp (complex!)
Full Algorithm
Animation • • • • Here's how you create an animated morph: Generate. Animation(Image 0, L 0[. . . ], Image 1, L 1[. . . ]) 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 L 0[i] to L 1[i]. end Warp 0 = Warp. Image( Image 0, L 0[. . . ], L[. . . ]) Warp 1 = Warp. Image( Image 1, L 1[. . . ], L[. . . ]) foreach pixel p in Final. Image do Final. Image(p) = (1 -t) Warp 0(p) + t Warp 1(p) end end
Interpolating Lines • Method 1: interpolating endpoints • Method 2: interpolating midpoints, length and orientation.
Results
Dynamic Scene
Algorithm summary
Morphing & matting • Extract foreground first to avoid artifacts in the background
Uniform morphing
Non-uniform morphing
Procedural Transformation
Multi-source morphing
Manipulating Facial Appearance through Shape and Color Duncan A. Rowland David I. Perrett St Andrews University IEEE CG&A, September 1995
The Morphable Face Model • shape vector S = (x 1, y 1, x 2, …, yn)T • appearance (texture) vector T = (R 1, G 1, B 1, R 2, …, Gn, Bn)T Shape S Appearance T
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
Face averaging by morphing average faces http: //www. beautycheck. de
Subpopulation means • Examples: – – – – Happy faces Young faces Asian faces Etc. Sunny days Rainy days Etc. Average female Average male
The average face
Women In Arts http: //www. youtube. com/watch? v=n. UDIo. N-_Hxs
References • Thaddeus Beier, Shawn Neely, Feature-Based Image Metamorphosis, SIGGRAPH 1992, pp 35 -42. • Detlef Ruprecht, Heinrich Muller, Image Warping with Scattered Data Interpolation, IEEE Computer Graphics and Applications, March 1995, pp 37 -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, pp 58 -71. • Peinsheng Gao, Thomas Sederberg, A work minimization approach to image morphing, The Visual Computer, 1998, pp 390 -400. • George Wolberg, Image morphing: a survey, The Visual Computer, 1998, pp 360 -372.
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
- Slides: 32