Deep learning project consist of 4 functionalities scratch removal, enhancing blurred image, noise removal and coloring images In this project, network that i used is called autoencoder. more specifically convolutional autoencoder.These kind of neural network tends perform better on imgaes, and thats how are we going to take advanates of autoencer as well as Convolution Nueral Network(CNN). Auto-encoders are used to recover using lossy data and CNN are used to achive good results while working with images. Thats why Convolutional Autoencoder is the best network for this project.
Dataset used: Labeled Faces in the Wild(LFW) http://vis-www.cs.umass.edu/lfw/
- Noise Removal
- Enhancing Blurred Image
- Scratch Removal
- Coloring Images
- Deployment Using Webapp
In any case if you need help feel free to contact me anytime