Colorization using Optimization Anat Levin Dani Lischinski Yair - - PowerPoint PPT Presentation
Colorization using Optimization Anat Levin Dani Lischinski Yair - - PowerPoint PPT Presentation
Colorization using Optimization Anat Levin Dani Lischinski Yair Weiss Colorization Colorization : a computer-assisted process of adding color to a monochrome image or ) movie. (Invented by Wilson Markle, 1970 Motivation Colorizing black
Colorization: a computer-assisted process of adding color to a monochrome image or
- movie. (Invented by Wilson Markle, 1970
)
Colorization
- Colorizing black and white movies / TV
shows
- Recoloring color images for special effects
Motivation
Color 101
- Typically represented by points in a three
dimensional color space:
- (Red, Green, Blue)
- Another option: [Y, U, V]
- Y - intensity
- U, V - chromatic channels
- A simple transformation exists:
[R, G, B] ⇔ [Y, U, V]
Typical Colorization Process
Images from: “Yet Another Colorization Tutorial” http://www.worth1000.com/tutorial.asp?sid= 161018
Images from: “Yet Another Colorization Tutorial” http://www.worth1000.com/tutorial.asp?sid= 161018
Typical Colorization Process
- Delineate region boundary
Images from: “Yet Another Colorization Tutorial” http://www.worth1000.com/tutorial.asp?sid= 161018
Typical Colorization Process
- Delineate region boundary
- Choose region color from palette
Images from: “Yet Another Colorization Tutorial” http://www.worth1000.com/tutorial.asp?sid= 161018
Typical Colorization Process
- Delineate region boundary
- Choose region color from palette
Images from: “Yet Another Colorization Tutorial” http://www.worth1000.com/tutorial.asp?sid= 161018
Typical Colorization Process
- Delineate region boundary
- Choose region color from palette
Process Limitations
- Requires expertise and time consuming
- Selecting detailed or fuzzy boundaries is
difficult
- Tracking regions is difficult (for video)
Let the user focus on the creative goals, without having to worry about selection!
Our Approach
Our Approach
Colors are propagated to every pixel in the image.
Grayscale channel Chroma channels
Pixel Affinity
Assumption: Neighboring pixels with similar intensities should have similar colors.
Pixel Affinity
Minimize:
r s wrs
Optimization
- Find a function U that minimizes:
(subject to the user’s color scribbles)
- Boils down to solving a system of linear
equations.
- Do the same for V.
Colorizing Stills
Original Colorized
Colorizing Stills
Colorizing Stills
13
- ut of 92 frames