search

jingyuanli001 / RFR-Inpainting

The source code for CVPR 2020 accepted paper "Recurrent Feature Reasoning for Image Inpainting"
MIT License
358 stars 76 forks source link

Question about the pretrain model #16

Closed ChujunWhu closed 4 years ago

ChujunWhu commented 4 years ago

Hi jingyuan, I would like to know the mask ratio of the pretrain model for CelebA. Have you used 40%-50% to train the model? And are the results under different mask ratio reported in your paper tested using a single pretrain model, or have you trained several models for different mask ratios. Thx!

jingyuanli001 commented 4 years ago

Hi jingyuan, I would like to know the mask ratio of the pretrain model for CelebA. Have you used 40%-50% to train the model? And are the results under different mask ratio reported in your paper tested using a single pretrain model, or have you trained several models for different mask ratios. Thx!

The results are from a single model trained with various mask ratios. The maximum mask ratio is ~60%.

jingyuanli001 commented 4 years ago

Hi jingyuan, I would like to know the mask ratio of the pretrain model for CelebA. Have you used 40%-50% to train the model? And are the results under different mask ratio reported in your paper tested using a single pretrain model, or have you trained several models for different mask ratios. Thx!

You might check the mask generation process described in partial convolution's paper. We followed the configurations in that paper.