dis-background-removal

Runtime error

App Files Files Community

soiz1 commited on 12 days ago

Commit

a9b3658

verified ·

1 Parent(s): 8df85b7

Update app.py

Browse files

Files changed (1) hide show

app.py +127 -176

app.py CHANGED Viewed

@@ -1,182 +1,133 @@
-<!DOCTYPE html>
-<html lang="en">
-<head>
-    <meta charset="UTF-8">
-    <meta name="viewport" content="width=device-width, initial-scale=1.0">
-    <title>Highly Accurate Dichotomous Image Segmentation</title>
-    <style>
-        body {
-            font-family: Arial, sans-serif;
-            max-width: 800px;
-            margin: 0 auto;
-            padding: 20px;
-            line-height: 1.6;
-        }
-        .container {
-            display: flex;
-            flex-direction: column;
-            gap: 20px;
-        }
-        .upload-section {
-            border: 2px dashed #ccc;
-            padding: 20px;
-            text-align: center;
-            border-radius: 5px;
-        }
-        .results {
-            display: flex;
-            gap: 20px;
-            flex-wrap: wrap;
-        }
-        .result-box {
-            flex: 1;
-            min-width: 300px;
-        }
-        img {
-            max-width: 100%;
-            height: auto;
-            border: 1px solid #ddd;
-            border-radius: 4px;
-        }
-        button {
-            background-color: #4CAF50;
-            color: white;
-            padding: 10px 15px;
-            border: none;
-            border-radius: 4px;
-            cursor: pointer;
-            font-size: 16px;
-        }
-        button:hover {
-            background-color: #45a049;
-        }
-        .code-block {
-            background-color: #f5f5f5;
-            padding: 15px;
-            border-radius: 5px;
-            overflow-x: auto;
-        }
-    </style>
-</head>
-<body>
-    <div class="container">
-        <h1>Highly Accurate Dichotomous Image Segmentation</h1>
-        <p>This is a demo for DIS, a model that can remove the background from a given image.</p>
-        <div class="upload-section">
-            <h2>Upload Image</h2>
-            <input type="file" id="imageInput" accept="image/*">
-            <button onclick="processImage()">Remove Background</button>
-        </div>
-        <div class="results">
-            <div class="result-box">
-                <h3>Original Image</h3>
-                <img id="originalImage" src="" alt="Original image will appear here" style="display: none;">
-            </div>
-            <div class="result-box">
-                <h3>Result (RGBA)</h3>
-                <img id="resultImage" src="" alt="Result will appear here" style="display: none;">
-            </div>
-            <div class="result-box">
-                <h3>Mask</h3>
-                <img id="maskImage" src="" alt="Mask will appear here" style="display: none;">
-            </div>
-        </div>
-        <div>
-            <h2>API Usage Example</h2>
-            <p>You can also use the API directly with this JavaScript code:</p>
-            <div class="code-block">
-                <pre><code>
-async function removeBackground(imageFile) {
-    const formData = new FormData();
-    formData.append('image', imageFile);
-    try {
-        const response = await fetch('/api/remove_bg', {
-            method: 'POST',
-            body: formData
-        });
-        if (!response.ok) {
-            throw new Error(`HTTP error! status: ${response.status}`);
-        }
-        const data = await response.json();
-        console.log('Result:', data);
-        return data;
-    } catch (error) {
-        console.error('Error:', error);
-        throw error;
-    }
-}
-// Usage example:
-// const fileInput = document.querySelector('input[type="file"]');
-// removeBackground(fileInput.files[0])
-//     .then(data => {
-//         // Handle response data
-//         document.getElementById('resultImage').src = data.rgba_url;
-//         document.getElementById('maskImage').src = data.mask_url;
-//     });
-                </code></pre>
-            </div>
-        </div>
-    </div>
-    <script>
-        function processImage() {
-            const fileInput = document.getElementById('imageInput');
-            if (!fileInput.files || fileInput.files.length === 0) {
-                alert('Please select an image first');
-                return;
-            }
-            const file = fileInput.files[0];
-            const reader = new FileReader();
-            reader.onload = function(e) {
-                document.getElementById('originalImage').src = e.target.result;
-                document.getElementById('originalImage').style.display = 'block';
-            };
-            reader.readAsDataURL(file);
-            removeBackground(file)
-                .then(data => {
-                    document.getElementById('resultImage').src = data.rgba_url;
-                    document.getElementById('resultImage').style.display = 'block';
-                    document.getElementById('maskImage').src = data.mask_url;
-                    document.getElementById('maskImage').style.display = 'block';
-                })
-                .catch(error => {
-                    console.error('Error:', error);
-                    alert('An error occurred while processing the image');
-                });
-        }
-        async function removeBackground(imageFile) {
-            const formData = new FormData();
-            formData.append('image', imageFile);
-            try {
-                const response = await fetch('/api/remove_bg', {
-                    method: 'POST',
-                    body: formData
-                });
-                if (!response.ok) {
-                    throw new Error(`HTTP error! status: ${response.status}`);
-                }
-                const data = await response.json();
-                console.log('Result:', data);
-                return data;
-            } catch (error) {
-                console.error('Error:', error);
-                throw error;
-            }
-        }
-    </script>
-</body>
-</html>

+from flask import Flask, request, jsonify, render_template
+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
+import uuid
+app = Flask(__name__)
+# モデル設定と初期化コード
+device = 'cuda' if torch.cuda.is_available() else 'cpu'
+class GOSNormalize(object):
+    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)
+        return image
+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)
+    im, im_shp = im_preprocess(im, hypar["cache_size"])
+    im = torch.divide(im,255.0)
+    shape = torch.from_numpy(np.array(im_shp))
+    return transform(im).unsqueeze(0), shape.unsqueeze(0)
+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):
+                layer.float()
+    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()
+    return net
+def predict(net, inputs_val, shapes_val, hypar, device):
+    net.eval()
+    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)
+    ds_val = net(inputs_val_v)[0]
+    pred_val = ds_val[0][0,:,:,:]
+    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)
+    if device == 'cuda': torch.cuda.empty_cache()
+    return (pred_val.detach().cpu().numpy()*255).astype(np.uint8)
+# モデル初期化
+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()
+}
+net = build_model(hypar, device)
+# 結果を保存するディレクトリを作成
+os.makedirs('static/results', exist_ok=True)
+@app.route('/')
+def index():
+    return render_template('index.html')
+@app.route('/api/remove_bg', methods=['POST'])
+def remove_bg():
+    if 'image' not in request.files:
+        return jsonify({'error': 'No image provided'}), 400
+    file = request.files['image']
+    if file.filename == '':
+        return jsonify({'error': 'No image selected'}), 400
+    # 一時ファイルとして保存
+    temp_path = f"static/temp_{uuid.uuid4().hex}.png"
+    file.save(temp_path)
+    try:
+        # 画像処理
+        image_tensor, orig_size = load_image(temp_path, hypar)
+        mask = predict(net, image_tensor, orig_size, hypar, device)
+        pil_mask = Image.fromarray(mask).convert('L')
+        im_rgb = Image.open(temp_path).convert("RGB")
+        # 結果を保存
+        result_id = uuid.uuid4().hex
+        rgba_path = f"static/results/{result_id}_rgba.png"
+        mask_path = f"static/results/{result_id}_mask.png"
+        im_rgba = im_rgb.copy()
+        im_rgba.putalpha(pil_mask)
+        im_rgba.save(rgba_path)
+        pil_mask.save(mask_path)
+        # 一時ファイルを削除
+        os.remove(temp_path)
+        return jsonify({
+            'rgba_url': f"/{rgba_path}",
+            'mask_url': f"/{mask_path}"
+        })
+    except Exception as e:
+        # エラーが発生したら一時ファイルを削除
+        if os.path.exists(temp_path):
+            os.remove(temp_path)
+        return jsonify({'error': str(e)}), 500
+if __name__ == '__main__':
+    app.run(debug=True)