upgraded luma aiu dream machine function call

lollms/functions/luma_ai_dream_machine/build_video_using_luma_ai.py

@@ -8,7 +8,7 @@ from pathlib import Path
 
 # Import necessary libraries
 from functools import partial
-from typing import Dict
+from typing import Dict, Tuple
 from lollms.utilities import PackageManager
 from ascii_colors import trace_exception
 
@@ -20,6 +20,38 @@ if not PackageManager.check_package_installed("pyautogui"):
 import pyautogui
 import webbrowser
 import time
+import numpy as np
+from lollms.utilities import PackageManager
+
+if not PackageManager.check_package_installed("cv2"):
+    PackageManager.install_package("opencv-python")
+import cv2
+from ascii_colors import ASCIIColors
+
+def capture_screenshot() -> np.ndarray:
+    """Capture a screenshot of the current screen."""
+    screenshot = pyautogui.screenshot()
+    screenshot = cv2.cvtColor(np.array(screenshot), cv2.COLOR_RGB2BGR)
+    return screenshot
+
+def template_matching(screenshot: np.ndarray, template_path: str, threshold: float = 0.8) -> Tuple[int, int, int, int]:
+    """Perform template matching to find the object in the screenshot."""
+    template = cv2.imread(str(template_path), cv2.IMREAD_COLOR)
+    template_gray = cv2.cvtColor(template, cv2.COLOR_BGR2GRAY)
+    screenshot_gray = cv2.cvtColor(screenshot, cv2.COLOR_BGR2GRAY)
+
+    result = cv2.matchTemplate(screenshot_gray, template_gray, cv2.TM_CCOEFF_NORMED)
+    min_val, max_val, min_loc, max_loc = cv2.minMaxLoc(result)
+
+    ASCIIColors.blue(f"Searching results:{max_val}")
+
+    if max_val >= threshold:
+        (startX, startY) = max_loc
+        endX = startX + template.shape[1]
+        endY = startY + template.shape[0]
+        return (startX, startY, endX, endY)
+    else:
+        return None
 
 def luma_ai_dream_machine_video_creator(prompt: str) -> str:
     """
@@ -36,22 +68,30 @@ def luma_ai_dream_machine_video_creator(prompt: str) -> str:
         # Open the Luma AI Dream Machine webpage
         webbrowser.open("https://lumalabs.ai/dream-machine/creations")
         time.sleep(2)  # Wait for the page to load
-
+        # Capture a screenshot of the browser window
+        screenshot = capture_screenshot()
         # Locate the input section and type the prompt
-        input_image_path = Path(__file__).parent/"input_section_image.png"  # Replace with the actual path to your image
-        if not input_image_path.exists():
+        template_path = Path(__file__).parent/"input_section_image.png"  # Replace with the actual path to your image
+        if not template_path.exists():
             raise FileNotFoundError("Input section image not found")
-
-        input_location = pyautogui.locateOnScreen(str(input_image_path))
+        # Perform template matching to find the object
+        match = template_matching(screenshot, template_path)
         
-        if input_location:
-            pyautogui.click(input_location)
+        if match:
+            startX, startY, endX, endY = match
+            print(f"Object found at ({startX}, {startY}) with width {endX - startX} and height {endY - startY}")
+            # Move the cursor to the center of the detected object
+            pyautogui.moveTo(startX + (endX - startX) // 2, startY + (endY - startY) // 2)
+            # Click the detected object
+            pyautogui.click()
+            # Type the specified text
             pyautogui.typewrite(prompt)
             pyautogui.press('enter')
-            return "Video creation in progress!"
+            return "Video generation process started. This may take some time"
         else:
             return "Please log in to Luma AI Dream Machine to create a video."
     except Exception as e:
+        trace_exception(e)
         return "Please log in to Luma AI Dream Machine to create a video."
 
 def luma_ai_dream_machine_video_creator_function() -> Dict: