Forensic Facial Identification using GANs

?Forensic Facial Identification using GANs? will provide the evidence in security investigation and criminal trials. Facial identification system uses facial features from a sketch or text based on eyewitness testimony. It compares the information with a database of known faces on w

Project Title

Forensic Facial Identification using GANs

Project Area of Specialization

Artificial Intelligence

Project Summary

“Forensic Facial Identification using GANs” will provide the evidence in security investigation and criminal trials. Facial identification system uses facial features from a sketch or text based on eyewitness testimony. It compares the information with a database of known faces on which it is trained to generate realistic images. Forensic facial identification goes beyond because it deals with facial features under unconstrained and non-ideal conditions, such as rough drawn sketches, varying facial orientation and a wide range of facial expressions. Images generated by this smart system will help security officers to identify the criminals easily. This smart system will be web-based.

Project Objectives

The main objective of this smart system is to help security officers to identify the criminals easily by using real-time Images generated by this system.

This smart system will be developed for “Security Officers” who are providing services in finding and investigating criminals. In any location through website.

We need such smart system because sometime may be security officer have to generate image on the sight of robbery to identify the criminal using testimony of eyewitness.

Project Implementation Method

This smart system will us Artificial Intllingance based GANs Machine Learrning Framework  to generate real-time images. GANs uses two neural network models in which  one model, the generator, acts akin to a painting forger. It tries to create images that look very similar to the dataset. The other model, the discriminator, acts like the police, and tries to detect whether the images generated were fake or not. GANs, or Generative Adversarial Networks, are a type of neural network architecture that allow neural networks to generate data. What basically happens, is that the generator keeps getting better at making fakes, while the discriminator keep getting better at detecting fakes. Effectively, these two models keep trying to beat each other, until after many iterations, the generator creates images indistinguishable from the real dataset.

Benefits of the Project

Forensic Facial Identification using GANs will provide evidence in police investigation and criminal trials. This smart system system will use the facial features from a sketch or text based on eyewitness’s testimony. Images generated by this smart system will help police officers to identify the criminals easily. This smart system will be web-based.

Technical Details of Final Deliverable

Our final product will be able to convert natural language text descriptions into images is an amazing demonstration of Deep Learning. Our final product will be capable that in the side of Generator network, the text embedding is filtered trough a fully connected layer and concatenated with the random noise vector z. In this case, the text embedding is converted from a 1024x1 vector to 128x1 and concatenated with the 100x1 random noise vector z. On the side of the discriminator network, the text-embedding is also compressed through a fully connected layer into a 128x1 vector and then reshaped into a 4x4 matrix and depth-wise concatenated with the image representation. This image representation is derived after the input image has been convolved over multiple times, reduce the spatial resolution and extracting information. This embedding strategy of Stack GANs for the discriminator is different from the conditional-GAN model in which the embedding is concatenated into the original image matrix and then convolved over.

Our final product will also be able to use the Sketch to Image generation using image-to-image translation model using Conditional Generative Adversarial Networks

Final Deliverable of the Project

Software System

Core Industry

Security

Other Industries

Core Technology

Artificial Intelligence(AI)

Other Technologies

Sustainable Development Goals

Industry, Innovation and Infrastructure

Required Resources

If you need this project, please contact me on contact@adikhanofficial.com