Suspect Face Detection by Auto Sketching

In the Modern world, the crime rate has been increasing day by day. To find the suspect face is a completely different task. So, the police want to draw the face sketch by the eye-witness. But it takes a lot of time and so to finish the sketch. Sometimes the face sketches are inaccurate because of the artist’s incorrect drawing and the eye-witness’s incorrect description. The main objective of the proposed work is to create an application using Generative Adversarial Networks (GAN) technology to generate unique images of human faces. our program will be generating more realistic images according to the inputs given by the user. This would be possible when the different slides of facial features such as Skin Tone, Facial Hair Length, Shape of the Head, etc are present in the application. These images are fake but it is similar to human faces. Our experimental outcomes are shown that the sketches generated by this application are more accurate and matched than the normal hand-drawn sketches. This application is directly used by the police to sketch persons. These sketches are better than hand draws due to the images created by using this application. It is convenient and more particle to use.