add divide param

models/SRFlow/code/__init__.py

@@ -35,7 +35,12 @@ def predict(model, lr):
     return visuals.get('rlt', visuals.get("SR"))
 
 
-def t(array): return torch.Tensor(np.expand_dims(array.transpose([2, 0, 1]), axis=0).astype(np.float32)) / 255
+def t(array, divide): 
+    output = torch.Tensor(np.expand_dims(array.transpose([2, 0, 1]), axis=0).astype(np.float32))
+    if divide:
+        return output / 255
+    else:
+        return output
 
 
 def rgb(t): return (

models/SRFlow/srflow.py

@@ -7,7 +7,7 @@ from PIL import Image
 import matplotlib.pyplot as plt
 from torchvision.transforms import PILToTensor, ToPILImage
 
-def return_SRFlow_result(lr, conf_path='models/SRFlow/code/confs/SRFlow_DF2K_4X.yml'):
+def return_SRFlow_result(lr, divide, conf_path='models/SRFlow/code/confs/SRFlow_DF2K_4X.yml'):
     """
     Apply Super-Resolution using SRFlow model to the input LR (low-resolution) image.
 
@@ -29,7 +29,7 @@ def return_SRFlow_result(lr, conf_path='models/SRFlow/code/confs/SRFlow_DF2K_4X.
     lr = impad(lr, bottom=int(np.ceil(h / pad_factor) * pad_factor - h),
                         right=int(np.ceil(w / pad_factor) * pad_factor - w))
     
-    lr_t = t(lr)
+    lr_t = t(lr, divide)
     heat = opt['heat']
     
     sr_t = model.get_sr(lq=lr_t, heat=heat)
@@ -40,7 +40,7 @@ def return_SRFlow_result(lr, conf_path='models/SRFlow/code/confs/SRFlow_DF2K_4X.
     sr = Image.fromarray((sr).astype('uint8'))
     return sr
 
-def return_SRFlow_result_from_tensor(lr_tensor):
+def return_SRFlow_result_from_tensor(lr_tensor, divide=True):
     """
     Apply Super-Resolution using SRFlow model to the input batched BCHW tensor.
 
@@ -55,7 +55,7 @@ def return_SRFlow_result_from_tensor(lr_tensor):
 
     for b in range(batch_size):
         lr_image = ToPILImage()(lr_tensor[b])
-        sr_image = return_SRFlow_result(lr_image)
+        sr_image = return_SRFlow_result(lr_image, divide)
         sr_tensor = PILToTensor()(sr_image).unsqueeze(0)
         sr_list.append(sr_tensor)