Apply for community grant: Academic project
This app demonstrates the editing capabilities of the White-box Style Transfer Editing (WISE) framework. It optimizes the parameters of classical image processing filters to match a given style image. After optimization, parameters can be tuned to achieve a desired look. Parameter tuning is done either globally or by painting the parameters locally.
The parameter style transfer demonstrated in this streamlit demo is based on our ECCV 2022 Paper "WISE: Whitebox Image Stylization by Example-based Learning". To show the editing capabilities, we would need access to a GPU grant, e.g. T4
Hi @MaxReimann , thanks for requesting, your demo and project look really cool. Very nice to see style transfer being made in a cool framework and real time edits. A T4 GPU is granted for your project on a temporary/use-based case!