Add macOS/Windows setup scripts and update modules for enhanced funct…

clone_or_update_deep_live_cam.bat

@@ -0,0 +1,20 @@
+@echo off
+REM clone_or_update_deep_live_cam.bat - Clone or update Deep-Live-Cam repo in a separate folder and sync to local working folder
+SET REPO_URL=https://github.com/hacksider/Deep-Live-Cam.git
+SET TARGET_DIR=Deep-Live-Cam-remote
+SET LOCAL_DIR=Deep-Live-Cam
+
+IF EXIST %TARGET_DIR% (
+    echo Updating existing repo in %TARGET_DIR% ...
+    cd %TARGET_DIR%
+    git pull
+    cd ..
+) ELSE (
+    echo Cloning repo to %TARGET_DIR% ...
+    git clone %REPO_URL% %TARGET_DIR%
+)
+
+REM Sync updated code to local working folder (excluding .git and models)
+xcopy %TARGET_DIR% %LOCAL_DIR% /E /H /Y /EXCLUDE:exclude.txt
+
+echo Done. Latest code is in %LOCAL_DIR%.

clone_or_update_deep_live_cam.sh

@@ -0,0 +1,20 @@
+#!/bin/zsh
+# clone_or_update_deep_live_cam.sh - Clone or update Deep-Live-Cam repo in a separate folder (macOS/Linux)
+REPO_URL="https://github.com/hacksider/Deep-Live-Cam.git"
+TARGET_DIR="Deep-Live-Cam-remote"
+
+if [ -d "$TARGET_DIR" ]; then
+  echo "Updating existing repo in $TARGET_DIR ..."
+  cd "$TARGET_DIR"
+  git pull
+  cd ..
+else
+  echo "Cloning repo to $TARGET_DIR ..."
+  git clone "$REPO_URL" "$TARGET_DIR"
+fi
+
+# Sync updated code to local working folder (excluding .git and models)
+LOCAL_DIR="Deep-Live-Cam"
+rsync -av --exclude='.git' --exclude='models' --exclude='*.pth' --exclude='*.onnx' "$TARGET_DIR"/ "$LOCAL_DIR"/
+
+echo "Done. Latest code is in $LOCAL_DIR."

exclude.txt

@@ -0,0 +1,4 @@
+.git
+models
+*.pth
+*.onnx

install_macos.sh

@@ -0,0 +1,44 @@
+#!/bin/bash
+# Deep-Live-Cam macOS Automated Setup
+set -e
+
+# 1. Ensure Homebrew is installed
+if ! command -v brew &> /dev/null; then
+    echo "Homebrew not found. Please install Homebrew first: https://brew.sh/"
+    exit 1
+fi
+
+# 2. Install Python 3.10 and tkinter
+brew install [email protected] [email protected]
+
+# 3. Create and activate virtual environment
+PYTHON_BIN=$(brew --prefix [email protected])/bin/python3.10
+$PYTHON_BIN -m venv venv
+source venv/bin/activate
+
+# 4. Upgrade pip and install dependencies
+pip install --upgrade pip
+pip install -r requirements.txt
+
+# 5. Download models if not present
+mkdir -p models
+if [ ! -f models/GFPGANv1.4.pth ]; then
+    curl -L -o models/GFPGANv1.4.pth "https://huggingface.co/hacksider/deep-live-cam/resolve/main/GFPGANv1.4.pth"
+fi
+if [ ! -f models/inswapper_128_fp16.onnx ]; then
+    curl -L -o models/inswapper_128_fp16.onnx "https://huggingface.co/hacksider/deep-live-cam/resolve/main/inswapper_128_fp16.onnx"
+fi
+
+# 6. Run instructions for user
+
+echo "\nSetup complete!"
+echo "To activate your environment and run Deep-Live-Cam, use one of the following commands:" 
+echo ""
+echo "# For CUDA (Nvidia GPU, if supported):"
+echo "source venv/bin/activate && python run.py --execution-provider cuda"
+echo ""
+echo "# For Apple Silicon (M1/M2/M3) CoreML:"
+echo "source venv/bin/activate && python3.10 run.py --execution-provider coreml"
+echo ""
+echo "# For CPU only:"
+echo "source venv/bin/activate && python run.py"

install_windows.bat

@@ -0,0 +1,36 @@
+@echo off
+REM Deep-Live-Cam Windows Automated Setup
+
+REM 1. Create virtual environment
+python -m venv venv
+if errorlevel 1 (
+    echo Failed to create virtual environment. Ensure Python 3.10+ is installed and in PATH.
+    exit /b 1
+)
+
+REM 2. Activate virtual environment
+call venv\Scripts\activate
+if errorlevel 1 (
+    echo Failed to activate virtual environment.
+    exit /b 1
+)
+
+REM 3. Install dependencies
+pip install --upgrade pip
+pip install -r requirements.txt
+if errorlevel 1 (
+    echo Failed to install dependencies.
+    exit /b 1
+)
+
+REM 4. Download models (manual step if not present)
+echo Downloading models (if not already in models/)...
+if not exist models\GFPGANv1.4.pth (
+    powershell -Command "Invoke-WebRequest -Uri https://huggingface.co/hacksider/deep-live-cam/resolve/main/GFPGANv1.4.pth -OutFile models\GFPGANv1.4.pth"
+)
+if not exist models\inswapper_128_fp16.onnx (
+    powershell -Command "Invoke-WebRequest -Uri https://huggingface.co/hacksider/deep-live-cam/resolve/main/inswapper_128_fp16.onnx -OutFile models\inswapper_128_fp16.onnx"
+)
+
+REM 5. Run the app
+python run.py

modules/capturer.py

@@ -4,29 +4,35 @@
 
 
 def get_video_frame(video_path: str, frame_number: int = 0) -> Any:
+    """Extract a specific frame from a video file, with color correction if enabled."""
     capture = cv2.VideoCapture(video_path)
-
-    # Set MJPEG format to ensure correct color space handling
-    capture.set(cv2.CAP_PROP_FOURCC, cv2.VideoWriter_fourcc(*'MJPG'))
-
-    # Only force RGB conversion if color correction is enabled
-    if modules.globals.color_correction:
-        capture.set(cv2.CAP_PROP_CONVERT_RGB, 1)
-
-    frame_total = capture.get(cv2.CAP_PROP_FRAME_COUNT)
-    capture.set(cv2.CAP_PROP_POS_FRAMES, min(frame_total, frame_number - 1))
-    has_frame, frame = capture.read()
-
-    if has_frame and modules.globals.color_correction:
-        # Convert the frame color if necessary
-        frame = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
-
-    capture.release()
-    return frame if has_frame else None
+    try:
+        # Set MJPEG format to ensure correct color space handling
+        capture.set(cv2.CAP_PROP_FOURCC, cv2.VideoWriter_fourcc(*'MJPG'))
+        # Only force RGB conversion if color correction is enabled
+        if modules.globals.color_correction:
+            capture.set(cv2.CAP_PROP_CONVERT_RGB, 1)
+        frame_total = capture.get(cv2.CAP_PROP_FRAME_COUNT)
+        capture.set(cv2.CAP_PROP_POS_FRAMES, min(frame_total, frame_number - 1))
+        has_frame, frame = capture.read()
+        if has_frame and modules.globals.color_correction:
+            frame = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
+        return frame if has_frame else None
+    except Exception as e:
+        print(f"Error extracting video frame: {e}")
+        return None
+    finally:
+        capture.release()
 
 
 def get_video_frame_total(video_path: str) -> int:
+    """Return the total number of frames in a video file."""
     capture = cv2.VideoCapture(video_path)
-    video_frame_total = int(capture.get(cv2.CAP_PROP_FRAME_COUNT))
-    capture.release()
-    return video_frame_total
+    try:
+        video_frame_total = int(capture.get(cv2.CAP_PROP_FRAME_COUNT))
+        return video_frame_total
+    except Exception as e:
+        print(f"Error getting video frame total: {e}")
+        return 0
+    finally:
+        capture.release()

modules/cluster_analysis.py

@@ -1,32 +1,42 @@
 import numpy as np
 from sklearn.cluster import KMeans
 from sklearn.metrics import silhouette_score
-from typing import Any
+from typing import Any, List, Tuple
 
 
-def find_cluster_centroids(embeddings, max_k=10) -> Any:
+def find_cluster_centroids(embeddings: List[Any], max_k: int = 10) -> Any:
+    """Find optimal cluster centroids for a set of embeddings using KMeans."""
     inertia = []
     cluster_centroids = []
     K = range(1, max_k+1)
 
     for k in K:
-        kmeans = KMeans(n_clusters=k, random_state=0)
-        kmeans.fit(embeddings)
-        inertia.append(kmeans.inertia_)
-        cluster_centroids.append({"k": k, "centroids": kmeans.cluster_centers_})
+        try:
+            kmeans = KMeans(n_clusters=k, random_state=0)
+            kmeans.fit(embeddings)
+            inertia.append(kmeans.inertia_)
+            cluster_centroids.append({"k": k, "centroids": kmeans.cluster_centers_})
+        except Exception as e:
+            print(f"KMeans failed for k={k}: {e}")
+
+    if len(inertia) < 2:
+        return cluster_centroids[0]['centroids'] if cluster_centroids else []
 
     diffs = [inertia[i] - inertia[i+1] for i in range(len(inertia)-1)]
     optimal_centroids = cluster_centroids[diffs.index(max(diffs)) + 1]['centroids']
 
     return optimal_centroids
 
-def find_closest_centroid(centroids: list, normed_face_embedding) -> list:
+
+def find_closest_centroid(centroids: List[Any], normed_face_embedding: Any) -> Tuple[int, Any]:
+    """Find the index and value of the centroid closest to the given embedding."""
     try:
         centroids = np.array(centroids)
         normed_face_embedding = np.array(normed_face_embedding)
         similarities = np.dot(centroids, normed_face_embedding)
         closest_centroid_index = np.argmax(similarities)
-        
+
         return closest_centroid_index, centroids[closest_centroid_index]
-    except ValueError:
-        return None
+    except Exception as e:
+        print(f"Error in find_closest_centroid: {e}")
+        return -1, None