My biggest inspiration was to help people who are unable to spread their amazing knowledge to this world because they feel discomfort in revealing their identities. This will give them a chance to spread knowledge freely. I also wanted to contribute to the booming media and entertainment industry with a new innovation.
What it does
It is capable of generating a face of your choice based on a multitude of parameters like 'smiling', 'sunglasses'. Creativity is the limit here! So you can create a model of your choice. Then you can use the famous deep fake algorithm to morph the face you just generated into moving the way you want it to.
How I built it
It is a combination of open-source repositories and custom add ons to make it suitable for my purposes. I also used Python, TensorFlow and streamlit to make it a fully functional web app.
Challenges I ran into
combining the various parts of the app ensuring that nothing breaks and the user experience remains consistent and intact.
Accomplishments that I'm proud of
Making a web app out of the deepfake and face gan algorithms.
What I learned
I learnt a lot about GANs and generative models in general and also the role my projects plays in helping more creators come up.
What's next for SYNGAN
Scaling and deploying it so that the masses can use it.