Deep learning-based techniques are at the centre of the attentions of companies, researchers and even general public. The applications on the recognition of people, places and objects have demonstrated the power of these techniques showing unprecedented success rates.
One interesting application of neural networks is in what is called “style transfer”. This corresponds to get a picture or illustration and by using an exemplar from a specific painter the system produces a version of the input picture with close resemblance to the works of that painter.
This work is to be developed in collaboration with a Portuguese movie studio, and is devoted at the exploration of the use of neural networks for style transfer for images and paintings and explore its application in new domains like film colourisation or augmented reality.
Work Plan:
- Study of neural networks and deep learning fundamentals
- Development of a style transfer application for simple images
- Exploring scenario recognition based on existing frameworks
- Automated adjustment of style transfer parameters from recognised scenarios
- Validation with video sequences
- Validation in Augmented Reality (tentative)
- Demonstration
