KenanHuang
commented
3 months ago I think you just need to replace the decoder of the Autoencoder_KL with the larger decoder implemented in Asy-VAGAN. You can follow the instruction of the "inpaint_st.py" and understand what dose it do in the script. Well as the description in the bottom of this issue #3
Firstly, thank you so much for your open source code. It has to be said that this is an excellent piece of work. I recently ran the code for Paint By Example and did experience the issue of missing details described in your paper. I see in your paper that there is an experimental comparison with Paint By Example. However, the implementation of this script is not found in your Code base. So, can you open up the script for implementing Paint By Example in your article? Or could you send it to me for learning? Thanks again！！！