You are on page 1of 6

Final Year Project Proposal on

3D Face Recognition under pose variation


Supervised by Dr. YN Singh Associate Professor, Computer Science and Engineering, Institute of Engineering And Technology, Lucknow Proposed by Anubhav Shrivastava Roll number:1005210009 Final Year

Basic challenges

Ideally, if all the images are similar, the task to identify the face would be trivial.

An ideal image

Image taken with mobile phone; distorted face

Improper background

Not looking straight into the camera

Improper flash or improper lightening

Shadow on face

Too much glare on spectacles

Dark tinted spectacles

Poor resolution

Process design

RAD model of software development life cycle will be used. The concept of modularity design will be used, i.e. divide the project into independent sub modules, develop them and later integrate them.

The concept of cohesion and coupling will be used.

Module 1

Module 2

Module 4

Module 3

Modules of the software

Broadly speaking there shall be 5 steps in face recognition system:

input

sensing

segmentation

Feature extraction

Classification

Post processing

decision

All the phases of the phase recognition system will have a different algorithms. The idea is develop these modules separately and the later integrate them.

Programing specifications

The software will be targeted to run on windows operation system Language: C# Frame work: Microsoft .NET

Integrated development Environment: Microsoft visual studio


C# supports rich library in image processing and mathematical work

Software development timeline

The idea is to develop a software with the trivial algorithms available and then later enhance the efficiency of algorithms. The targeted timeline of the development of the project is :
Learning of programming in C# and .NET framework
By 31st December

Post processing algorithm


By 15st march

Integration of modules
30th April

Feature Extraction Algorithm


By 15th January

Sensing algorithm
By 28th march

Documentation
May

Feature classification algorithm


By 31st January

Segmentation Algorithm
By 15th February

You might also like