Variational Exemplar-Based Image Colorization

Abstract:-

Interest point detection is an important research area in the field of image

processing and computer vision. In particular, image retrieval and object

categorization heavily rely on interest. Point detection from which local image

descriptors are computed for image matching. The use of color may therefore provide

selective search reducing the total number of interest points used for image matching.

This paper proposes color interest points for sparse image representation. A large-

scale experiment, it is shown that the proposed color interest point detector has higher

repeatability than a luminance-based one. Furthermore, in the context of image

retrieval, a reduced and predictable number of color features show an increase in

performance compared to state-of-the-art interest points. The contributions of this

paper are twofold. First, we introduce a variational formulation.Modeling the color

selection problem under spatial constraints and propose a minimization scheme, which

computes a local Minima of the defined non convex energy. Second, we combine

GLOBALSOFT TECHNOLOGIESIEEE PROJECTS & SOFTWARE DEVELOPMENTS

IEEE FINAL YEAR PROJECTS|IEEE ENGINEERING PROJECTS|IEEE STUDENTS PROJECTS|IEEE

BULK PROJECTS|BE/BTECH/ME/MTECH/MS/MCA PROJECTS|CSE/IT/ECE/EEE PROJECTS

CELL: +91 98495 39085, +91 99662 35788, +91 98495 57908, +91 97014 40401

Visit: www.finalyearprojects.org Mail to:ieeefinalsemprojects@gmail.com

GLOBALSOFT TECHNOLOGIESIEEE PROJECTS & SOFTWARE DEVELOPMENTS

IEEE FINAL YEAR PROJECTS|IEEE ENGINEERING PROJECTS|IEEE STUDENTS PROJECTS|IEEE

BULK PROJECTS|BE/BTECH/ME/MTECH/MS/MCA PROJECTS|CSE/IT/ECE/EEE PROJECTS

CELL: +91 98495 39085, +91 99662 35788, +91 98495 57908, +91 97014 40401

Visit: www.finalyearprojects.org Mail to:ieeefinalsemprojects@gmail.com

different patch-based features and distances in order to construct a consistent set of

possible color candidates. A first category of colorization algorithms requires

a huge amount of user intervention to manually add initial colors to the grayscale

image. The colorization process is then performed by propagating the input color data

to the whole image. First, several color candidates are computed using different

features and associated metrics. The final color is then obtained by solving a

variational model which allows the automatic selection of the best candidates

while adding some regularization in the solution.

Existing System:-

The candidate selection and color regularization problems are

simultaneously solved through a variational energy

minimization.

Natural images contain different types of complex structures,

redundancies and textures.

The problem of selecting the best color to transfer among the

set of possible colors is solved, for each pixel independently,

using a dimensional reduction.

We focus on the image colorization problem which involves

consistent and spatially coherent image results.

Proposed System:

We explore methods for learning local image descriptors from training data We

describe a set of building blocks for constructing descriptors which can be

combined together.

We consider both linear and nonlinear transforms with dimensionality

reduction.

These techniques have state-of-the-art performance in all of our test scenarios;

have been used to design local feature descriptors for a robust structure.

The color interest point detectors are presented and a scale selection method is

proposed.

Hardware Requirements:-

SYSTEM : Pentium IV 2.4 GHz

HARD DISK : 40 GB

RAM : 256 MB

Software Requirements:-

Operating system : Windows XP Professional

IDE : Microsoft Visual Studio .Net 2005

Coding Language : C# .NET