moving width and height to varible and setting width and height to one given by shein

app.py

@@ -121,6 +121,9 @@ pipe = TryonPipeline.from_pretrained(
 )
 pipe.unet_encoder = UNet_Encoder
 
+WIDTH = 4160 # 768
+HEIGHT = 6240 # 1024
+
 @spaces.GPU
 def start_tryon(dict,garm_img,garment_des,is_checked,is_checked_crop,denoise_steps,seed):
     device = "cuda"
@@ -129,7 +132,7 @@ def start_tryon(dict,garm_img,garment_des,is_checked,is_checked_crop,denoise_ste
     pipe.to(device)
     pipe.unet_encoder.to(device)
 
-    garm_img= garm_img.convert("RGB").resize((768,1024))
+    garm_img= garm_img.convert("RGB").resize((WIDTH,HEIGHT))
     human_img_orig = dict["background"].convert("RGB")    
     
     if is_checked_crop:
@@ -142,25 +145,25 @@ def start_tryon(dict,garm_img,garment_des,is_checked,is_checked_crop,denoise_ste
         bottom = (height + target_height) / 2
         cropped_img = human_img_orig.crop((left, top, right, bottom))
         crop_size = cropped_img.size
-        human_img = cropped_img.resize((768,1024))
+        human_img = cropped_img.resize((WIDTH, HEIGHT))
     else:
-        human_img = human_img_orig.resize((768,1024))
+        human_img = human_img_orig.resize((WIDTH, HEIGHT))
 
 
     if is_checked:
-        keypoints = openpose_model(human_img.resize((384,512)))
-        model_parse, _ = parsing_model(human_img.resize((384,512)))
+        keypoints = openpose_model(human_img.resize((WIDTH/2,HEIGHT/2)))
+        model_parse, _ = parsing_model(human_img.resize((WIDTH/2,HEIGHT/2)))
         mask, mask_gray = get_mask_location('hd', "upper_body", model_parse, keypoints)
-        mask = mask.resize((768,1024))
+        mask = mask.resize((WIDTH,HEIGHT))
     else:
-        mask = pil_to_binary_mask(dict['layers'][0].convert("RGB").resize((768, 1024)))
+        mask = pil_to_binary_mask(dict['layers'][0].convert("RGB").resize((WIDTH, HEIGHT)))
         # mask = transforms.ToTensor()(mask)
         # mask = mask.unsqueeze(0)
     mask_gray = (1-transforms.ToTensor()(mask)) * tensor_transfrom(human_img)
     mask_gray = to_pil_image((mask_gray+1.0)/2.0)
 
 
-    human_img_arg = _apply_exif_orientation(human_img.resize((384,512)))
+    human_img_arg = _apply_exif_orientation(human_img.resize((WIDTH/2, HEIGHT/2)))
     human_img_arg = convert_PIL_to_numpy(human_img_arg, format="BGR")
      
     
@@ -169,7 +172,7 @@ def start_tryon(dict,garm_img,garment_des,is_checked,is_checked_crop,denoise_ste
     # verbosity = getattr(args, "verbosity", None)
     pose_img = args.func(args,human_img_arg)    
     pose_img = pose_img[:,:,::-1]    
-    pose_img = Image.fromarray(pose_img).resize((768,1024))
+    pose_img = Image.fromarray(pose_img).resize((WIDTH,HEIGHT))
     
     with torch.no_grad():
         # Extract the images
@@ -227,9 +230,9 @@ def start_tryon(dict,garm_img,garment_des,is_checked,is_checked_crop,denoise_ste
                         cloth = garm_tensor.to(device,torch.float16),
                         mask_image=mask,
                         image=human_img, 
-                        height=1024,
-                        width=768,
-                        ip_adapter_image = garm_img.resize((768,1024)),
+                        height=HEIGHT,
+                        width=WIDTH,
+                        ip_adapter_image = garm_img.resize((WIDTH,HEIGHT)),
                         guidance_scale=2.0,
                     )[0]