Update app.py

app.py

@@ -132,9 +132,8 @@ def save_image(img):
     img.save(unique_name, format="WEBP", lossless=True) 
     return unique_name
 
-#@spaces.GPU
-# Assurez-vous que les fonctions de traitement utilisent asyncio
-async def start_tryon(dict, garm_img, garment_des, is_checked, is_checked_crop, denoise_steps, seed, categorie='upper_body'):
+@spaces.GPU
+def start_tryon(dict, garm_img, garment_des, is_checked, is_checked_crop, denoise_steps, seed, categorie = 'upper_body'):
     device = "cuda"
     openpose_model.preprocessor.body_estimation.model.to(device)
     pipe.to(device)
@@ -158,27 +157,20 @@ async def start_tryon(dict, garm_img, garment_des, is_checked, is_checked_crop,
         human_img = human_img_orig.resize((768, 1024))
 
     if is_checked:
-        # Utilisation de asyncio pour rendre le calcul non bloquant
-        keypoints = await asyncio.to_thread(openpose_model, human_img.resize((384, 512)))
-        model_parse, _ = await asyncio.to_thread(parsing_model, human_img.resize((384, 512)))
-        mask, mask_gray = await asyncio.to_thread(get_mask_location, 'hd', categorie, model_parse, keypoints)
+        keypoints = openpose_model(human_img.resize((384, 512)))
+        model_parse, _ = parsing_model(human_img.resize((384, 512)))
+        mask, mask_gray = get_mask_location('hd', categorie , model_parse, keypoints)
         mask = mask.resize((768, 1024))
     else:
-        mask = dict['layers'][0].convert("RGB").resize((768, 1024))
-
+        mask = dict['layers'][0].convert("RGB").resize((768, 1024))#pil_to_binary_mask(dict['layers'][0].convert("RGB").resize((768, 1024)))
     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 = convert_PIL_to_numpy(human_img_arg, format="BGR")
 
-    # Création des arguments de façon non bloquante
-    args = await asyncio.to_thread(
-        apply_net.create_argument_parser().parse_args,
-        ('show', './configs/densepose_rcnn_R_50_FPN_s1x.yaml', './ckpt/densepose/model_final_162be9.pkl', 'dp_segm', '-v', '--opts', 'MODEL.DEVICE', 'cuda')
-    )
-
-    pose_img = await asyncio.to_thread(args.func, args, human_img_arg)
+    args = apply_net.create_argument_parser().parse_args(('show', './configs/densepose_rcnn_R_50_FPN_s1x.yaml', './ckpt/densepose/model_final_162be9.pkl', 'dp_segm', '-v', '--opts', 'MODEL.DEVICE', 'cuda'))
+    pose_img = args.func(args, human_img_arg)
     pose_img = pose_img[:, :, ::-1]
     pose_img = Image.fromarray(pose_img).resize((768, 1024))
 
@@ -221,10 +213,7 @@ async def start_tryon(dict, garm_img, garment_des, is_checked, is_checked_crop,
                 pose_img = tensor_transfrom(pose_img).unsqueeze(0).to(device, torch.float16)
                 garm_tensor = tensor_transfrom(garm_img).unsqueeze(0).to(device, torch.float16)
                 generator = torch.Generator(device).manual_seed(seed) if seed is not None else None
-                
-                # L'appel principal au modèle doit également être asynchrone
-                images = await asyncio.to_thread(
-                    pipe,
+                images = pipe(
                     prompt_embeds=prompt_embeds.to(device, torch.float16),
                     negative_prompt_embeds=negative_prompt_embeds.to(device, torch.float16),
                     pooled_prompt_embeds=pooled_prompt_embeds.to(device, torch.float16),
@@ -248,12 +237,13 @@ async def start_tryon(dict, garm_img, garment_des, is_checked, is_checked_crop,
         human_img_orig.paste(out_img, (int(left), int(top)))
         return human_img_orig, mask_gray
     else:
-        return images[0], mask_gray, mask
-
+        return images[0], mask_gray , mask
 
+        
 @app.route('/tryon-v2', methods=['POST'])
-async def tryon_v2():
-    data = await request.json
+def tryon_v2():
+
+    data = request.json
     human_image_data = data['human_image']
     garment_image_data = data['garment_image']
 
@@ -278,13 +268,11 @@ async def tryon_v2():
         'composite': None
     }
 
-    # Exécuter le traitement principal de manière asynchrone
-    output_image, mask_image, mask = await start_tryon(human_dict, garment_image, description, use_auto_mask, use_auto_crop, denoise_steps, seed, categorie)
-    
+    output_image, mask_image , mask = start_tryon(human_dict, garment_image, description, use_auto_mask, use_auto_crop, denoise_steps, seed, categorie)
     return jsonify({
         'image_id': save_image(output_image),
-        'mask_gray_id': save_image(mask_image),
-        'mask_id': save_image(mask)
+        'mask_gray_id' : save_image(mask_image),
+        'mask_id' : save_image(mask)
     })
 
 def clear_gpu_memory():