Create ui.py

ui.py

@@ -0,0 +1,435 @@
+"""
+Gradio UI für AI Image Generator - Mit Mock-Funktionen für CPU-Testing
+"""
+import gradio as gr
+from PIL import Image, ImageDraw
+import random
+import time
+
+# ==================== MOCK-FUNKTIONEN ====================
+def mock_text_to_image(prompt, model_id, steps, guidance_scale, progress=None):
+    """Mock für Text-zu-Bild: Erzeugt einfaches Farbbild"""
+    print(f"📝 Mock text_to_image aufgerufen: '{prompt[:50]}...'")
+    
+    if progress:
+        for i in range(10):  # Simuliere Fortschritt
+            time.sleep(0.05)
+            progress((i+1)/10, desc="Mock-Generierung läuft...")
+    
+    # Erstelle einfaches Farbbild
+    colors = ["lightblue", "lightgreen", "lavender", "peachpuff"]
+    img = Image.new('RGB', (512, 512), color=random.choice(colors))
+    
+    # Füge Text hinzu (simulierter Prompt)
+    draw = ImageDraw.Draw(img)
+    text = f"Mock: {prompt[:30]}..." if len(prompt) > 30 else f"Mock: {prompt}"
+    draw.text((50, 256), text, fill="black")
+    
+    status = f"✅ Mock: '{prompt[:30]}...' würde mit {model_id} generiert (Steps: {steps}, CFG: {guidance_scale})"
+    return img, status
+
+def mock_img_to_image(image, prompt, neg_prompt, strength, steps, guidance_scale,
+                     mode, bbox_x1, bbox_y1, bbox_x2, bbox_y2, progress=None):
+    """Mock für Bild-zu-Bild: Fügt Rahmen und Text hinzu"""
+    print(f"🎨 Mock img_to_image aufgerufen: Modus={mode}, Prompt='{prompt[:30]}...'")
+    
+    if not image:
+        return None, None, None, None, None
+    
+    if progress:
+        for i in range(10):
+            time.sleep(0.05)
+            progress((i+1)/10, desc=f"Mock-{mode} läuft...")
+    
+    # Kopiere Bild und füge Effekte hinzu
+    img = image.copy().convert("RGB")
+    draw = ImageDraw.Draw(img)
+    
+    # Rahmen basierend auf Modus
+    colors = {
+        "environment_change": "green",
+        "focus_change": "orange", 
+        "face_only_change": "red"
+    }
+    border_color = colors.get(mode, "blue")
+    
+    # Zeichne Rahmen
+    draw.rectangle([20, 20, img.width-20, img.height-20], 
+                   outline=border_color, width=10)
+    
+    # BBox visualisieren (falls angegeben)
+    if all(v is not None for v in [bbox_x1, bbox_y1, bbox_x2, bbox_y2]):
+        x1, y1 = min(bbox_x1, bbox_x2), min(bbox_y1, bbox_y2)
+        x2, y2 = max(bbox_x1, bbox_x2), max(bbox_y1, bbox_y2)
+        draw.rectangle([x1, y1, x2, y2], outline="yellow", width=5)
+        draw.text((x1+5, y1+5), f"{mode}", fill="white")
+    
+    # Text hinzufügen
+    draw.text((30, 30), f"Mock: {mode}", fill=border_color)
+    draw.text((30, 60), f"Strength: {strength}", fill="black")
+    
+    # Mock-Masken (einfache Platzhalter)
+    mask_preview = Image.new('RGB', (256, 256), color='gray')
+    controlnet_map = Image.new('RGB', (256, 256), color='darkblue')
+    canny_map = Image.new('RGB', (256, 256), color='black')
+    
+    status = f"✅ Mock: {mode} angewendet (Strength: {strength}, Steps: {steps})"
+    print(status)
+    
+    return img, mask_preview, mask_preview, controlnet_map, canny_map
+
+def mock_update_live_preview(image, bbox_x1, bbox_y1, bbox_x2, bbox_y2, mode):
+    """Mock für Live-Vorschau"""
+    if not image:
+        return None
+    
+    preview = image.copy()
+    draw = ImageDraw.Draw(preview)
+    
+    if all(v is not None for v in [bbox_x1, bbox_y1, bbox_x2, bbox_y2]):
+        x1, y1 = min(bbox_x1, bbox_x2), min(bbox_y1, bbox_y2)
+        x2, y2 = max(bbox_x1, bbox_x2), max(bbox_y1, bbox_y2)
+        
+        colors = {"environment_change": "green", 
+                 "focus_change": "orange", 
+                 "face_only_change": "red"}
+        
+        draw.rectangle([x1, y1, x2, y2], outline=colors.get(mode, "blue"), width=3)
+        draw.text((x1+5, y1-20), f"Mock: {mode}", fill=colors.get(mode, "blue"))
+    
+    return preview
+
+def mock_process_image_upload(image):
+    """Mock für Bild-Upload: Setzt BBox in der Mitte"""
+    if not image:
+        return None, 100, 100, 300, 300
+    
+    w, h = image.size
+    bbox_size = min(w, h) * 0.3
+    x1 = (w - bbox_size) / 2
+    y1 = (h - bbox_size) / 4
+    x2 = x1 + bbox_size
+    y2 = y1 + bbox_size * 1.2
+    
+    preview = mock_update_live_preview(image, x1, y1, x2, y2, "environment_change")
+    return preview, int(x1), int(y1), int(x2), int(y2)
+
+def mock_update_slider_for_image(image):
+    """Mock für Slider-Update"""
+    if not image:
+        return (
+            gr.update(maximum=4096),
+            gr.update(maximum=4096),
+            gr.update(maximum=4096),
+            gr.update(maximum=4096)
+        )
+    
+    w, h = image.size
+    max_val = min(max(w, h), 4096)
+    
+    print(f"📐 Mock Slider-Maxima: {max_val}")
+    return (
+        gr.update(maximum=max_val),
+        gr.update(maximum=max_val),
+        gr.update(maximum=max_val),
+        gr.update(maximum=max_val)
+    )
+
+def mock_update_model_settings(model_id):
+    """Mock für Modelleinstellungen"""
+    configs = {
+        "runwayml/stable-diffusion-v1-5": (35, 7.5, "🏠 SD 1.5 Mock"),
+        "SG161222/Realistic_Vision_V6.0_B1_noVAE": (40, 7.0, "👤 Realistic Vision Mock")
+    }
+    steps, cfg, msg = configs.get(model_id, (35, 7.5, "Standard Mock"))
+    return steps, cfg, f"📊 Mock-Einstellungen: {msg}"
+
+def mock_update_info(mode):
+    """Mock für Prompt-Info-Update"""
+    if mode == "environment_change":
+        return (
+            "`[STIL-MOTIV], [UMGEBUNG], [VOR/HINTERGRUND]`",
+            "`[GESICHTER/ANATOMIE], [FEHLER]`"
+        )
+    elif mode == "focus_change":
+        return (
+            "`[GESICHTSBESCHREIBUNG], [KLEIDUNG], [POSITION]`",
+            "`[DEFORMIERT], [UNSCHÄRFE], [ANATOMIEFEHLER]`"
+        )
+    else:
+        return (
+            "`[HAARFARBE], [AUGEN], [GESICHTSAUSDRUCK]`",
+            "`[UNREALISTISCH], [ASYMETRISCH]`"
+        )
+
+# ==================== UI-DEFINITION ====================
+def main_ui():
+    """Haupt-UI-Funktion (angepasst für 3 Modi)"""
+    with gr.Blocks(
+        title="AI Image Generator - Mock Version",
+        theme=gr.themes.Base(),
+        css="""
+        .info-box { background: #f8fafc; padding: 8px 12px; border-radius: 6px; margin-bottom: 6px; }
+        .model-info-box { background: #e8f4fd; padding: 12px; border-radius: 6px; margin: 10px 0; border-left: 4px solid #2196f3; }
+        #generate-button { background-color: #0080FF !important; margin: 20px auto !important; width: 280px; }
+        .radio-group { background: #f8f9fa; padding: 15px; border-radius: 8px; margin: 10px 0; }
+        """
+    ) as demo:
+        
+        with gr.Tab("Text zu Bild"):
+            gr.Markdown("## 🎨 Text zu Bild Generator (Mock Version)")
+            
+            with gr.Row():
+                with gr.Column(scale=2):
+                    model_dropdown = gr.Dropdown(
+                        choices=[
+                            ("🏠 Stable Diffusion 1.5 (Mock)", "runwayml/stable-diffusion-v1-5"),
+                            ("👤 Realistic Vision V6.0 (Mock)", "SG161222/Realistic_Vision_V6.0_B1_noVAE")
+                        ],
+                        value="runwayml/stable-diffusion-v1-5",
+                        label="📁 Modellauswahl"
+                    )
+                    
+                    model_info_box = gr.Markdown(
+                        value="<div class='model-info-box'><strong>🏠 Stable Diffusion 1.5 (Mock)</strong><br>Universal model, good all-rounder<br><em>Mock-Einstellungen: 35 Steps, CFG 7.5</em></div>",
+                        label="Modellinformationen"
+                    )
+                    
+                with gr.Column(scale=3):
+                    txt_input = gr.Textbox(
+                        placeholder="z.B. ultra realistic mountain landscape at sunrise...",
+                        lines=3,
+                        label="🎯 Prompt (Englisch)",
+                        value="beautiful landscape sunset mountains"
+                    )
+            
+            with gr.Row():
+                with gr.Column():
+                    txt_steps = gr.Slider(
+                        minimum=10, maximum=100, value=35, step=1,
+                        label="⚙️ Inferenz-Schritte"
+                    )
+                with gr.Column():
+                    txt_guidance = gr.Slider(
+                        minimum=1.0, maximum=20.0, value=7.5, step=0.5,
+                        label="🎛️ Prompt-Stärke (CFG Scale)"
+                    )
+            
+            status_output = gr.Markdown(
+                value="",
+                elem_classes="status-message"
+            )
+            
+            generate_btn = gr.Button("🚀 Mock-Bild generieren", variant="primary", elem_id="generate-button")
+            
+            with gr.Row():
+                txt_output = gr.Image(
+                    label="🖼️ Generiertes Mock-Bild", 
+                    show_download_button=True,
+                    type="pil",
+                    height=400
+                )
+            
+            # Event-Handler
+            model_dropdown.change(
+                fn=mock_update_model_settings,
+                inputs=[model_dropdown],
+                outputs=[txt_steps, txt_guidance, model_info_box]
+            )
+            
+            generate_btn.click(
+                fn=mock_text_to_image,
+                inputs=[txt_input, model_dropdown, txt_steps, txt_guidance],
+                outputs=[txt_output, status_output],
+                concurrency_limit=1
+            )
+        
+        with gr.Tab("Bild zu Bild"):
+            gr.Markdown("## 🖼️ Bild zu Bild Transformation (Mock Version)")
+            
+            with gr.Row():
+                with gr.Column():
+                    img_input = gr.Image(
+                        type="pil", 
+                        label="📤 Eingabebild",
+                        height=300,
+                        sources=["upload"],
+                        elem_id="image-upload"
+                    )
+                with gr.Column():
+                    preview_output = gr.Image(
+                        label="🎯 Mock-Live-Vorschau",
+                        height=300,
+                        interactive=False
+                    )
+            
+            with gr.Row():
+                with gr.Column():
+                    gr.Markdown("### 🎛️ Transformations-Modus")
+                    mode_radio = gr.Radio(
+                        choices=[
+                            ("🌳 Umgebung ändern", "environment_change"),
+                            ("🎯 Focus verändern", "focus_change"),
+                            ("👤 Ausschließlich Gesicht", "face_only_change")
+                        ],
+                        value="environment_change",
+                        label="Wähle den Transformationsmodus:",
+                        elem_classes="radio-group"
+                    )
+            
+            with gr.Row():
+                gr.Markdown("### 📐 Bildelementbereich anpassen")
+            
+            with gr.Row():
+                with gr.Column():
+                    bbox_x1 = gr.Slider(
+                        label="← Links (x1)", 
+                        minimum=0, maximum=4096, value=100, step=1
+                    )
+                with gr.Column():
+                    bbox_y1 = gr.Slider(
+                        label="↑ Oben (y1)", 
+                        minimum=0, maximum=4096, value=100, step=1
+                    )
+            with gr.Row():
+                with gr.Column():
+                    bbox_x2 = gr.Slider(
+                        label="→ Rechts (x2)", 
+                        minimum=0, maximum=4096, value=300, step=1
+                    )
+                with gr.Column():
+                    bbox_y2 = gr.Slider(
+                        label="↓ Unten (y2)", 
+                        minimum=0, maximum=4096, value=300, step=1
+                    )
+            
+            with gr.Row():
+                with gr.Column():
+                    pos_info = gr.Markdown(
+                        value="`[STIL-MOTIV], [UMGEBUNG], [VOR/HINTERGRUND]`",
+                        elem_classes=["info-box"]
+                    )
+                    img_prompt = gr.Textbox(
+                        placeholder="photorealistic coastal beach, keep person unchanged...",
+                        lines=2,
+                        label="🎯 Transformations-Prompt",
+                        value="change background to beach",
+                        elem_classes=["prompt-box"]
+                    )
+                with gr.Column():
+                    neg_info = gr.Markdown(
+                        value="`[GESICHTER/ANATOMIE], [FEHLER]`",
+                        elem_classes=["info-box"]
+                    )        
+                    img_neg_prompt = gr.Textbox(
+                        placeholder="blurry face, deformed anatomy...",
+                        lines=2,
+                        label="🚫 Negativ-Prompt",
+                        value="blurry, ugly",
+                        elem_classes=["prompt-box"]
+                    )
+            
+            with gr.Row():
+                with gr.Column():
+                    strength_slider = gr.Slider(
+                        minimum=0.1, maximum=0.9, value=0.4, step=0.05,
+                        label="💪 Veränderungs-Stärke"
+                    )
+                with gr.Column():
+                    img_steps = gr.Slider(
+                        minimum=10, maximum=45, value=35, step=1,
+                        label="⚙️ Inferenz-Schritte"
+                    )
+                with gr.Column():
+                    img_guidance = gr.Slider(
+                        minimum=1.0, maximum=15.0, value=7.5, step=0.5,
+                        label="🎛️ Prompt-Stärke"
+                    )
+            
+            gr.Markdown("### 📋 Hinweise:\n• **Mock-Version** für UI-Testing\n• **Keine echte KI-Verarbeitung**\n• **Teste alle UI-Elemente**")
+            
+            transform_btn = gr.Button("🔄 Mock-Bild transformieren", variant="primary")
+            
+            with gr.Row():
+                img_output = gr.Image(
+                    label="✨ Transformiertes Mock-Bild",
+                    show_download_button=True,
+                    type="pil",
+                    height=400
+                )
+            
+            with gr.Row():
+                sam_raw_mask_output = gr.Image(
+                    label="🔍 SAM-Rohmaske (Mock)",
+                    type="pil",
+                    height=250,
+                    show_download_button=False
+                )
+                processed_mask_output = gr.Image(
+                    label="🛠️ Nachbearbeitete Maske (Mock)",
+                    type="pil",
+                    height=250,
+                    show_download_button=False
+                )
+            
+            with gr.Row():                 
+                pose_map_output = gr.Image(
+                    label="🎭 Pose/Depth Map (Mock)", 
+                    type="pil", 
+                    height=250,
+                    show_download_button=False
+                )
+                canny_map_output = gr.Image(
+                    label="📐 Canny Edge Map (Mock)", 
+                    type="pil", 
+                    height=250,
+                    show_download_button=False
+                )
+            
+            # Event-Handler
+            img_input.upload(
+                fn=mock_process_image_upload,
+                inputs=[img_input],
+                outputs=[preview_output, bbox_x1, bbox_y1, bbox_x2, bbox_y2]
+            ).then(
+                fn=mock_update_slider_for_image,
+                inputs=[img_input],
+                outputs=[bbox_x1, bbox_y1, bbox_x2, bbox_y2]
+            )
+            
+            coordinate_inputs = [img_input, bbox_x1, bbox_y1, bbox_x2, bbox_y2, mode_radio]
+            
+            for slider in [bbox_x1, bbox_y1, bbox_x2, bbox_y2]:
+                slider.release(
+                    fn=mock_update_live_preview,
+                    inputs=coordinate_inputs,
+                    outputs=preview_output
+                )
+            
+            mode_radio.change(
+                fn=mock_update_info,
+                inputs=[mode_radio],
+                outputs=[pos_info, neg_info]
+            ).then(
+                fn=mock_update_live_preview,
+                inputs=coordinate_inputs,
+                outputs=preview_output
+            )
+            
+            transform_btn.click(
+                fn=mock_img_to_image,
+                inputs=[
+                    img_input, img_prompt, img_neg_prompt, 
+                    strength_slider, img_steps, img_guidance, 
+                    mode_radio, bbox_x1, bbox_y1, bbox_x2, bbox_y2
+                ],
+                outputs=[img_output, sam_raw_mask_output, processed_mask_output, 
+                        pose_map_output, canny_map_output],
+                concurrency_limit=1
+            )
+    
+    return demo
+
+if __name__ == "__main__":
+    demo = main_ui()
+    demo.launch(server_name="0.0.0.0", server_port=7860)