Python Image Processing Cookbook

How to do it...

Execute the following steps to complete the missing parts of an image:

1. Define the tensor2image() and image2tensor() functions to convert from a PyTorch tensor into an OpenCV-Python image and back, using the following code snippet:

def tensor2image(src): out = src.copy() * 255 out = out.transpose((1, 2, 0)).astype(np.uint8) out = cv2.cvtColor(out, cv2.COLOR_RGB2BGR) return outdef image2tensor(src): out = src.copy() out = cv2.cvtColor(out, cv2.COLOR_BGR2RGB) out = out.transpose((2,0,1)).astype(np.float64) / 255 return out

2. Define the input and mask image paths and the model path. Check whether the GPU is available:

image_path = 'images/zebra.png'mask_path = 'images/inpaint_mask.png'model_path = 'models/completionnet_places2.t7' ...