Hi,guy。I read your article "EdgeConnect: Generative Image Inpainting with Adversarial Edge Learning". I have some questions. When you training, you using 'mask' parameter prevents canny to detect edges for the masked regions. So how to get the mask when repairing the real picture?
Hi,guy。I read your article "EdgeConnect: Generative Image Inpainting with Adversarial Edge Learning". I have some questions. When you training, you using 'mask' parameter prevents canny to detect edges for the masked regions. So how to get the mask when repairing the real picture?