Create app.py

app.py

@@ -0,0 +1,443 @@
+import gradio as gr
+import torch
+import numpy as np
+from PIL import Image
+import cv2
+
+print("🚀 Starting SAM2 App v2.0...")
+
+device = "cuda" if torch.cuda.is_available() else "cpu"
+print(f"📱 Using device: {device}")
+
+model = None
+processor = None
+
+def load_model():
+    global model, processor
+    if model is None:
+        print("📦 Loading SAM model...")
+        try:
+            from transformers import SamModel, SamProcessor
+            
+            model_name = "facebook/sam-vit-large"
+            
+            processor = SamProcessor.from_pretrained(model_name)
+            model = SamModel.from_pretrained(model_name)
+            model.to(device)
+            print(f"✅ Model loaded: {model_name}")
+        except Exception as e:
+            print(f"❌ Error: {e}, falling back to base model")
+            model_name = "facebook/sam-vit-base"
+            processor = SamProcessor.from_pretrained(model_name)
+            model = SamModel.from_pretrained(model_name)
+            model.to(device)
+    return model, processor
+
+def prepare_image(image, max_size=1024):
+    if isinstance(image, np.ndarray):
+        image_pil = Image.fromarray(image)
+    else:
+        image_pil = image
+    
+    if image_pil.mode != 'RGB':
+        image_pil = image_pil.convert('RGB')
+    
+    image_np = np.array(image_pil)
+    h, w = image_np.shape[:2]
+    
+    if max(h, w) > max_size:
+        scale = max_size / max(h, w)
+        new_h, new_w = int(h * scale), int(w * scale)
+        image_pil = image_pil.resize((new_w, new_h), Image.Resampling.LANCZOS)
+        image_np = np.array(image_pil)
+    
+    return image_pil, image_np
+
+def refine_mask(mask, kernel_size=5):
+    mask_uint8 = (mask > 0).astype(np.uint8) * 255
+    kernel = cv2.getStructuringElement(cv2.MORPH_ELLIPSE, (kernel_size, kernel_size))
+    mask_closed = cv2.morphologyEx(mask_uint8, cv2.MORPH_CLOSE, kernel)
+    mask_refined = cv2.morphologyEx(mask_closed, cv2.MORPH_OPEN, kernel)
+    return mask_refined > 0
+
+def merge_nearby_masks(masks, scores, distance_threshold=100):
+    """
+    Kombiniert Masken die nah beieinander sind
+    → Fisch + Flosse werden EINS!
+    """
+    if len(masks) == 0:
+        return np.zeros_like(masks[0]) if len(masks) > 0 else None
+    
+    # Finde Zentren aller Masken
+    centers = []
+    for mask in masks:
+        y_coords, x_coords = np.where(mask)
+        if len(x_coords) > 0:
+            center_x = int(np.mean(x_coords))
+            center_y = int(np.mean(y_coords))
+            centers.append((center_x, center_y))
+        else:
+            centers.append(None)
+    
+    # Gruppiere Masken die nah beieinander sind
+    combined = np.zeros_like(masks[0], dtype=bool)
+    
+    for i, mask in enumerate(masks):
+        if centers[i] is None:
+            continue
+        
+        # Checke ob diese Maske nah an bereits kombinierten Masken ist
+        if np.sum(combined) > 0:
+            # Distanz zwischen Masken
+            combined_coords = np.argwhere(combined)
+            mask_coords = np.argwhere(mask)
+            
+            if len(combined_coords) > 0 and len(mask_coords) > 0:
+                # Minimale Distanz zwischen den Masken
+                min_dist = float('inf')
+                for mc in mask_coords[::10]:  # Sample alle 10 Punkte
+                    for cc in combined_coords[::10]:
+                        dist = np.linalg.norm(mc - cc)
+                        min_dist = min(min_dist, dist)
+                
+                # Wenn nah genug: kombinieren!
+                if min_dist < distance_threshold:
+                    combined = combined | mask
+                    print(f"  ✅ Merged mask {i} (dist: {min_dist:.1f})")
+        else:
+            # Erste Maske
+            combined = mask
+    
+    return combined
+
+def segment_automatic(image, quality="high", merge_parts=True):
+    """
+    Automatische Segmentierung mit Multi-Point Detection
+    merge_parts=True → Kombiniert getrennte Teile wie Flossen!
+    """
+    if image is None:
+        return None, {"error": "Kein Bild hochgeladen"}
+    
+    try:
+        print(f"🔄 Starting segmentation (quality: {quality}, merge: {merge_parts})...")
+        model, processor = load_model()
+        
+        image_pil, image_np = prepare_image(image)
+        h, w = image_np.shape[:2]
+        
+        # MEHRERE Punkte für bessere Abdeckung!
+        center_x, center_y = w // 2, h // 2
+        points = [
+            [center_x, center_y],           # Mitte
+            [center_x, center_y - h//4],    # Oben
+            [center_x, center_y + h//4],    # Unten
+            [center_x - w//4, center_y],    # Links
+            [center_x + w//4, center_y]     # Rechts
+        ]
+        
+        all_masks = []
+        all_scores = []
+        
+        print(f"🎯 Using {len(points)} detection points...")
+        
+        for idx, point in enumerate(points):
+            inputs = processor(
+                image_pil,
+                input_points=[[point]],
+                input_labels=[[1]],
+                return_tensors="pt"
+            ).to(device)
+            
+            with torch.no_grad():
+                outputs = model(**inputs, multimask_output=True)
+            
+            masks = processor.image_processor.post_process_masks(
+                outputs.pred_masks.cpu(),
+                inputs["original_sizes"].cpu(),
+                inputs["reshaped_input_sizes"].cpu()
+            )[0]
+            
+            scores = outputs.iou_scores.cpu().numpy()
+            if scores.ndim > 1:
+                scores = scores.flatten()
+            
+            # Beste Maske von diesem Punkt
+            best_idx = np.argmax(scores)
+            
+            if masks.ndim == 4:
+                mask = masks[0, best_idx].numpy()
+            else:
+                mask = masks[best_idx].numpy()
+            
+            # Nur gute Masken
+            if scores[best_idx] > 0.6:
+                all_masks.append(mask > 0)
+                all_scores.append(scores[best_idx])
+                print(f"  ✅ Point {idx}: Score {scores[best_idx]:.3f}")
+        
+        print(f"✅ Got {len(all_masks)} quality masks")
+        
+        # MERGE: Kombiniere nahe Masken
+        if merge_parts and len(all_masks) > 0:
+            combined_mask = merge_nearby_masks(all_masks, all_scores, distance_threshold=100)
+        else:
+            combined_mask = np.zeros((h, w), dtype=bool)
+            for mask in all_masks:
+                combined_mask = combined_mask | mask
+        
+        # Refinement
+        if quality == "high":
+            combined_mask = refine_mask(combined_mask, kernel_size=9)
+        
+        # Overlay
+        overlay = image_np.copy()
+        color = np.array([255, 80, 180])  # Rosa/Pink
+        
+        mask_float = combined_mask.astype(float)
+        if quality == "high":
+            mask_float = cv2.GaussianBlur(mask_float, (7, 7), 0)
+        
+        for c in range(3):
+            overlay[:, :, c] = (
+                overlay[:, :, c] * (1 - mask_float * 0.65) +
+                color[c] * mask_float * 0.65
+            )
+        
+        # Kontur
+        contours, _ = cv2.findContours(
+            combined_mask.astype(np.uint8),
+            cv2.RETR_EXTERNAL,
+            cv2.CHAIN_APPROX_SIMPLE
+        )
+        cv2.drawContours(overlay, contours, -1, (255, 255, 0), 3)
+        
+        metadata = {
+            "success": True,
+            "mode": "automatic_plus" if merge_parts else "automatic",
+            "quality": quality,
+            "detection_points": len(points),
+            "masks_found": len(all_masks),
+            "avg_score": float(np.mean(all_scores)) if all_scores else 0,
+            "all_scores": [float(s) for s in all_scores],
+            "image_size": [w, h],
+            "mask_area": int(np.sum(combined_mask)),
+            "mask_percentage": float(np.sum(combined_mask) / (h * w) * 100),
+            "num_contours": len(contours),
+            "device": device
+        }
+        
+        print(f"✅ Segmentation complete! {len(contours)} contours")
+        return Image.fromarray(overlay.astype(np.uint8)), metadata
+        
+    except Exception as e:
+        import traceback
+        print(f"❌ ERROR:\n{traceback.format_exc()}")
+        return image, {"error": str(e)}
+
+def segment_multi_dense(image, density="medium"):
+    """Multi-Object mit dichterem Grid"""
+    if image is None:
+        return None, {"error": "Kein Bild"}
+    
+    try:
+        model, processor = load_model()
+        image_pil, image_np = prepare_image(image)
+        h, w = image_np.shape[:2]
+        
+        if density == "high":
+            grid_size = 5
+        elif density == "medium":
+            grid_size = 4
+        else:
+            grid_size = 3
+        
+        points = []
+        for i in range(1, grid_size + 1):
+            for j in range(1, grid_size + 1):
+                x = int(w * i / (grid_size + 1))
+                y = int(h * j / (grid_size + 1))
+                points.append([x, y])
+        
+        print(f"📍 Using {len(points)} points for segmentation")
+        
+        all_masks = []
+        all_scores = []
+        
+        for idx, point in enumerate(points):
+            inputs = processor(
+                image_pil,
+                input_points=[[point]],
+                input_labels=[[1]],
+                return_tensors="pt"
+            ).to(device)
+            
+            with torch.no_grad():
+                outputs = model(**inputs, multimask_output=True)
+            
+            masks = processor.image_processor.post_process_masks(
+                outputs.pred_masks.cpu(),
+                inputs["original_sizes"].cpu(),
+                inputs["reshaped_input_sizes"].cpu()
+            )[0]
+            
+            scores = outputs.iou_scores.cpu().numpy().flatten()
+            best_idx = np.argmax(scores)
+            
+            if masks.ndim == 4:
+                mask = masks[0, best_idx].numpy()
+            else:
+                mask = masks[best_idx].numpy()
+            
+            if scores[best_idx] > 0.7:
+                all_masks.append(refine_mask(mask))
+                all_scores.append(scores[best_idx])
+        
+        print(f"✅ Got {len(all_masks)} quality masks")
+        
+        overlay = image_np.copy()
+        
+        colors = []
+        for i in range(len(all_masks)):
+            hue = int(180 * i / max(len(all_masks), 1))
+            color_hsv = np.uint8([[[hue, 255, 200]]])
+            color_rgb = cv2.cvtColor(color_hsv, cv2.COLOR_HSV2RGB)[0][0]
+            colors.append(color_rgb)
+        
+        for mask, color, score in zip(all_masks, colors, all_scores):
+            alpha = 0.4 + (score - 0.7) * 0.2
+            overlay[mask] = (
+                overlay[mask] * (1 - alpha) +
+                np.array(color) * alpha
+            ).astype(np.uint8)
+            
+            contours, _ = cv2.findContours(
+                mask.astype(np.uint8),
+                cv2.RETR_EXTERNAL,
+                cv2.CHAIN_APPROX_SIMPLE
+            )
+            cv2.drawContours(overlay, contours, -1, color.tolist(), 2)
+        
+        metadata = {
+            "success": True,
+            "mode": "multi_object_dense",
+            "density": density,
+            "grid_size": f"{grid_size}x{grid_size}",
+            "total_points": len(points),
+            "quality_masks": len(all_masks),
+            "avg_score": float(np.mean(all_scores)) if all_scores else 0,
+            "scores": [float(s) for s in all_scores]
+        }
+        
+        return Image.fromarray(overlay), metadata
+        
+    except Exception as e:
+        import traceback
+        print(f"❌ ERROR:\n{traceback.format_exc()}")
+        return image, {"error": str(e)}
+
+# Gradio Interface
+demo = gr.Blocks(title="SAM2 Boostly", theme=gr.themes.Soft())
+
+with demo:
+    gr.Markdown("# 🎨 SAM2 Segmentierung - Boostly Edition")
+    gr.Markdown("### Zero-Shot Object Segmentation powered by Meta SAM")
+    
+    with gr.Tab("🤖 Automatisch PLUS"):
+        gr.Markdown("**Intelligent Multi-Point Detection** - Erkennt und kombiniert alle Objektteile!")
+        gr.Markdown("✨ Perfekt für: Fische mit Flossen, Menschen mit Armen/Beinen, komplexe Objekte")
+        
+        with gr.Row():
+            with gr.Column():
+                input_auto = gr.Image(type="pil", label="📸 Bild hochladen")
+                
+                quality_radio = gr.Radio(
+                    choices=["high", "fast"],
+                    value="high",
+                    label="Qualität"
+                )
+                
+                merge_checkbox = gr.Checkbox(
+                    value=True,
+                    label="✅ Teile zusammenfügen (Empfohlen!)"
+                )
+                
+                btn_auto = gr.Button("🚀 Segmentieren", variant="primary", size="lg")
+                
+                gr.Markdown("""
+                **So funktioniert's:**
+                - 🎯 5 Erkennungspunkte (Mitte + 4 Richtungen)
+                - 🔗 Intelligentes Merging naher Bereiche
+                - 🎨 Fisch + Flosse = EIN zusammenhängendes Objekt!
+                """)
+            
+            with gr.Column():
+                output_auto = gr.Image(label="✨ Ergebnis")
+                json_auto = gr.JSON(label="📊 Details")
+        
+        btn_auto.click(
+            fn=segment_automatic,
+            inputs=[input_auto, quality_radio, merge_checkbox],
+            outputs=[output_auto, json_auto]
+        )
+    
+    with gr.Tab("🎯 Multi-Region"):
+        gr.Markdown("**Grid-basierte Segmentierung** - Für mehrere separate Objekte")
+        
+        with gr.Row():
+            with gr.Column():
+                input_multi = gr.Image(type="pil", label="📸 Bild hochladen")
+                density_radio = gr.Radio(
+                    choices=["high", "medium", "low"],
+                    value="medium",
+                    label="Punkt-Dichte"
+                )
+                btn_multi = gr.Button("🎯 Alle Bereiche", variant="primary", size="lg")
+                
+                gr.Markdown("""
+                **Grid-Größen:**
+                - 🔥 High: 5x5 = 25 Punkte
+                - ⚡ Medium: 4x4 = 16 Punkte  
+                - 💨 Low: 3x3 = 9 Punkte
+                """)
+            
+            with gr.Column():
+                output_multi = gr.Image(label="✨ Ergebnis")
+                json_multi = gr.JSON(label="📊 Details")
+        
+        btn_multi.click(
+            fn=segment_multi_dense,
+            inputs=[input_multi, density_radio],
+            outputs=[output_multi, json_multi]
+        )
+    
+    with gr.Tab("📡 API Integration"):
+        gr.Markdown("### API Endpoint für Lovable/externe Apps")
+        gr.Code(
+            "https://EnginDev-Boostly.hf.space/api/predict",
+            label="API URL"
+        )
+        
+        gr.Markdown("### JavaScript Beispiel")
+        gr.Code('''
+const response = await fetch('https://EnginDev-Boostly.hf.space/api/predict', {
+    method: 'POST',
+    headers: {'Content-Type': 'application/json'},
+    body: JSON.stringify({
+        data: [imageBase64, "high", true],  // [image, quality, merge]
+        fn_index: 0  // 0 = Automatisch PLUS, 1 = Multi-Region
+    })
+});
+const result = await response.json();
+const segmentedImage = result.data[0];  // Base64 Image
+const metadata = result.data[1];        // JSON Details
+        ''', language="javascript")
+        
+        gr.Markdown("### Parameter")
+        gr.Markdown("- **fn_index**: 0 = Automatisch PLUS, 1 = Multi-Region")
+        gr.Markdown("- **quality**: 'high' (präzise) oder 'fast' (schnell)")
+        gr.Markdown("- **merge**: true (Teile kombinieren) oder false")
+        gr.Markdown("- **density**: 'high', 'medium', 'low' (nur Multi-Region)")
+
+if __name__ == "__main__":
+    print("🌐 Launching Boostly SAM2...")
+    demo.launch(server_name="0.0.0.0", server_port=7860, show_error=True)