Update app.py

app.py

@@ -1,12 +1,11 @@
-import cv2
 import os
-from PIL import Image
 import numpy as np
 import torch
 from torch.autograd import Variable
 from torchvision import transforms
 import torch.nn.functional as F
 from flask import Flask, request, jsonify, render_template, send_from_directory
+from PIL import Image
 import warnings
 warnings.filterwarnings("ignore")
 
@@ -24,39 +23,48 @@ device = 'cuda' if torch.cuda.is_available() else 'cpu'
 
 # Download official weights if not exists
 if not os.path.exists("saved_models"):
-    os.mkdir("saved_models")
+    os.makedirs("saved_models", exist_ok=True)
     if not os.path.exists("saved_models/isnet.pth"):
-        os.system("mv isnet.pth saved_models/")
+        if os.path.exists("isnet.pth"):
+            os.rename("isnet.pth", "saved_models/isnet.pth")
 
 class GOSNormalize(object):
-    '''
-    Normalize the Image using torch.transforms
-    '''
-    def __init__(self, mean=[0.485,0.456,0.406], std=[0.229,0.224,0.225]):
+    def __init__(self, mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]):
         self.mean = mean
         self.std = std
 
-    def __call__(self,image):
-        image = normalize(image,self.mean,self.std)
+    def __call__(self, image):
+        image = normalize(image, self.mean, self.std)
         return image
 
-transform = transforms.Compose([GOSNormalize([0.5,0.5,0.5],[1.0,1.0,1.0])])
+transform = transforms.Compose([GOSNormalize([0.5, 0.5, 0.5], [1.0, 1.0, 1.0])])
 
 def load_image(im_path, hypar):
     im = im_reader(im_path)
-    # Convert to RGB if image has alpha channel
-    if im.mode == 'RGBA':
-        im = im.convert('RGB')
+    
+    # Convert numpy array to PIL Image if needed
+    if isinstance(im, np.ndarray):
+        if im.ndim == 3 and im.shape[2] == 4:  # RGBA image
+            im = Image.fromarray(im).convert('RGB')
+        elif im.ndim == 3:  # RGB image
+            im = Image.fromarray(im)
+        elif im.ndim == 2:  # Grayscale image
+            im = Image.fromarray(im).convert('RGB')
+    
+    # If it's already PIL Image, check mode
+    elif hasattr(im, 'mode'):
+        if im.mode == 'RGBA':
+            im = im.convert('RGB')
+    
     im, im_shp = im_preprocess(im, hypar["cache_size"])
-    im = torch.divide(im,255.0)
+    im = torch.divide(im, 255.0)
     shape = torch.from_numpy(np.array(im_shp))
-    return transform(im).unsqueeze(0), shape.unsqueeze(0) # make a batch of image, shape
-
-def build_model(hypar,device):
-    net = hypar["model"]#GOSNETINC(3,1)
+    return transform(im).unsqueeze(0), shape.unsqueeze(0)
 
-    # convert to half precision
-    if(hypar["model_digit"]=="half"):
+def build_model(hypar, device):
+    net = hypar["model"]
+    
+    if hypar["model_digit"] == "half":
         net.half()
         for layer in net.modules():
             if isinstance(layer, nn.BatchNorm2d):
@@ -64,49 +72,48 @@ def build_model(hypar,device):
 
     net.to(device)
 
-    if(hypar["restore_model"]!=""):
-        net.load_state_dict(torch.load(hypar["model_path"]+"/"+hypar["restore_model"], map_location=device))
-        net.to(device)
-    net.eval()  
+    if hypar["restore_model"] != "":
+        net.load_state_dict(torch.load(os.path.join(hypar["model_path"], hypar["restore_model"]), 
+                                     map_location=device))
+    net.eval()
     return net
 
 def predict(net, inputs_val, shapes_val, hypar, device):
-    '''
-    Given an Image, predict the mask
-    '''
     net.eval()
 
-    if(hypar["model_digit"]=="full"):
+    if hypar["model_digit"] == "full":
         inputs_val = inputs_val.type(torch.FloatTensor)
     else:
         inputs_val = inputs_val.type(torch.HalfTensor)
 
-    inputs_val_v = Variable(inputs_val, requires_grad=False).to(device) # wrap inputs in Variable
-    ds_val = net(inputs_val_v)[0] # list of 6 results
+    inputs_val_v = Variable(inputs_val, requires_grad=False).to(device)
+    ds_val = net(inputs_val_v)[0]
 
-    pred_val = ds_val[0][0,:,:,:] # B x 1 x H x W    # we want the first one which is the most accurate prediction
+    pred_val = ds_val[0][0, :, :, :]
 
-    ## recover the prediction spatial size to the orignal image size
-    pred_val = torch.squeeze(F.upsample(torch.unsqueeze(pred_val,0),(shapes_val[0][0],shapes_val[0][1]),mode='bilinear'))
+    pred_val = torch.squeeze(F.upsample(torch.unsqueeze(pred_val, 0), 
+                           (shapes_val[0][0], shapes_val[0][1]), mode='bilinear'))
 
     ma = torch.max(pred_val)
     mi = torch.min(pred_val)
-    pred_val = (pred_val-mi)/(ma-mi) # max = 1
+    pred_val = (pred_val - mi) / (ma - mi)
 
-    if device == 'cuda': torch.cuda.empty_cache()
-    return (pred_val.detach().cpu().numpy()*255).astype(np.uint8) # it is the mask we need
+    if device == 'cuda': 
+        torch.cuda.empty_cache()
+    return (pred_val.detach().cpu().numpy() * 255).astype(np.uint8)
 
 # Set Parameters
-hypar = {} # paramters for inferencing
-hypar["model_path"] ="./saved_models" ## load trained weights from this path
-hypar["restore_model"] = "isnet.pth" ## name of the to-be-loaded weights
-hypar["interm_sup"] = False ## indicate if activate intermediate feature supervision
-hypar["model_digit"] = "full" ## indicates "half" or "full" accuracy of float number
-hypar["seed"] = 0
-hypar["cache_size"] = [1024, 1024] ## cached input spatial resolution
-hypar["input_size"] = [1024, 1024] ## model input spatial size
-hypar["crop_size"] = [1024, 1024] ## random crop size from the input
-hypar["model"] = ISNetDIS()
+hypar = {
+    "model_path": "./saved_models",
+    "restore_model": "isnet.pth",
+    "interm_sup": False,
+    "model_digit": "full",
+    "seed": 0,
+    "cache_size": [1024, 1024],
+    "input_size": [1024, 1024],
+    "crop_size": [1024, 1024],
+    "model": ISNetDIS()
+}
 
 # Build Model
 net = build_model(hypar, device)
@@ -171,4 +178,4 @@ def serve_upload(filename):
     return send_from_directory(app.config['UPLOAD_FOLDER'], filename)
 
 if __name__ == '__main__':
-    app.run(host='0.0.0.0', port=7860, debug=True)
+    app.run(host='0.0.0.0', port=5000, debug=True)