utils.py

import numpy as np
import cv2
from PIL import Image, ImageEnhance
import tempfile
import os
from moviepy.editor import ImageClip, AudioFileClip, CompositeVideoClip
import librosa
import soundfile as sf
import subprocess

def process_image_for_video(image, style="static"):
    """
    Process image for video creation based on style
    
    Args:
        image: PIL Image
        style: Animation style
    
    Returns:
        str: Path to processed image
    """
    # Convert to RGB if necessary
    if image.mode != 'RGB':
        image = image.convert('RGB')
    
    # Enhance image quality
    enhancer = ImageEnhance.Contrast(image)
    image = enhancer.enhance(1.2)
    
    enhancer = ImageEnhance.Sharpness(image)
    image = enhancer.enhance(1.1)
    
    # Save processed image
    temp_path = os.path.join(tempfile.gettempdir(), f"processed_{uuid.uuid4().hex}.jpg")
    image.save(temp_path, quality=95)
    
    return temp_path

def extend_audio_to_duration(audio_path, target_duration):
    """
    Extend audio to target duration by looping and fading
    
    Args:
        audio_path: Path to original audio file
        target_duration: Target duration in seconds
    
    Returns:
        str: Path to extended audio file
    """
    # Load audio
    y, sr = librosa.load(audio_path)
    
    original_duration = len(y) / sr
    
    if original_duration >= target_duration:
        return audio_path
    
    # Calculate how many times to loop
    loops_needed = int(target_duration / original_duration) + 1
    
    # Create extended audio
    extended_audio = np.tile(y, loops_needed)
    
    # Trim to exact duration
    target_samples = int(target_duration * sr)
    extended_audio = extended_audio[:target_samples]
    
    # Apply fade in/out
    fade_samples = int(0.5 * sr)  # 0.5 second fade
    extended_audio[:fade_samples] *= np.linspace(0, 1, fade_samples)
    extended_audio[-fade_samples:] *= np.linspace(1, 0, fade_samples)
    
    # Save extended audio
    output_path = os.path.join(tempfile.gettempdir(), f"extended_{uuid.uuid4().hex}.wav")
    sf.write(output_path, extended_audio, sr)
    
    return output_path

def create_video_with_audio(image_path, audio_path, duration):
    """
    Create video by combining image and audio
    
    Args:
        image_path: Path to image file
        audio_path: Path to audio file
        duration: Video duration in seconds
    
    Returns:
        str: Path to output video file
    """
    # Create video clip from image
    image_clip = ImageClip(image_path, duration=duration)
    
    # Create audio clip
    audio_clip = AudioFileClip(audio_path)
    
    # Set audio to video
    video_clip = image_clip.set_audio(audio_clip)
    
    # Set fps
    video_clip.fps = 24
    
    # Output path
    output_path = os.path.join(tempfile.gettempdir(), f"video_{uuid.uuid4().hex}.mp4")
    
    # Write video file
    video_clip.write_videofile(
        output_path,
        codec='libx264',
        audio_codec='aac',
        temp_audiofile=os.path.join(tempfile.gettempdir(), f"temp_audio_{uuid.uuid4().hex}.m4a"),
        remove_temp=True
    )
    
    # Close clips
    image_clip.close()
    audio_clip.close()
    video_clip.close()
    
    return output_path

def cleanup_temp_files(file_paths):
    """Clean up temporary files"""
    for file_path in file_paths:
        try:
            if os.path.exists(file_path):
                os.remove(file_path)
        except Exception as e:
            print(f"Error removing {file_path}: {e}")

def apply_video_effects(image_path, style, duration):
    """
    Apply video effects to create animated video from static image
    
    Args:
        image_path: Path to image
        style: Animation style
        duration: Duration in seconds
    
    Returns:
        str: Path to animated video
    """
    # Read image
    image = cv2.imread(image_path)
    height, width = image.shape[:2]
    
    # Create temporary video path
    temp_video = os.path.join(tempfile.gettempdir(), f"animated_{uuid.uuid4().hex}.mp4")
    
    # Setup video writer
    fourcc = cv2.VideoWriter_fourcc(*'mp4v')
    out = cv2.VideoWriter(temp_video, fourcc, 30.0, (width, height))
    
    frames = int(duration * 30)
    
    for i in range(frames):
        progress = i / frames
        
        if style == "zoom_in":
            # Zoom in effect
            scale = 1.0 + (progress * 0.5)
            new_height = int(height / scale)
            new_width = int(width / scale)
            y_start = (height - new_height) // 2
            x_start = (width - new_width) // 2
            frame = image[y_start:y_start+new_height, x_start:x_start+new_width]
            frame = cv2.resize(frame, (width, height))
            
        elif style == "zoom_out":
            # Zoom out effect
            scale = 1.5 - (progress * 0.5)
            new_height = int(height / scale)
            new_width = int(width / scale)
            y_start = (height - new_height) // 2
            x_start = (width - new_width) // 2
            frame = image[y_start:y_start+new_height, x_start:x_start+new_width]
            frame = cv2.resize(frame, (width, height))
            
        elif style == "pan_left":
            # Pan left effect
            offset = int(progress * (width * 0.3))
            frame = image[:, offset:offset+width] if offset+width <= width else image[:, -width:]
            
        elif style == "pan_right":
            # Pan right effect
            offset = int((1 - progress) * (width * 0.3))
            frame = image[:, offset:offset+width] if offset+width <= width else image[:, -width:]
            
        else:  # static
            frame = image
        
        out.write(frame)
    
    out.release()
    return temp_video