image_utils.py

import io
import urllib.request
import cv2
import numpy as np
from PIL import Image
from config import (
    MAX_DIM, MIN_DIM, SQUARE_DIM, MULTIPLE_OF, 
    FIXED_FPS, MIN_FRAMES_MODEL, MAX_FRAMES_MODEL
)

def load_image_from_url(url: str) -> Image.Image:
    if not url or not str(url).strip():
        raise ValueError("Masukkan URL gambar terlebih dahulu.")
    url = str(url).strip()
    headers = {"User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64)"}
    req = urllib.request.Request(url, headers=headers)
    with urllib.request.urlopen(req, timeout=15) as resp:
        img_bytes = resp.read()
    img = Image.open(io.BytesIO(img_bytes))
    return img.convert("RGB")

get_timestamp_js = """
function(video, timestamp) {
    const videoElem = document.querySelector('#generated-video video');
    let currentTime = 0;
    if (videoElem) {
        currentTime = videoElem.currentTime;
        console.log("Video found! Time: " + currentTime);
    } else {
        console.log("No video element found.");
    }
    return [video, currentTime];
}
"""

def extract_frame(video_path, timestamp):
    if not video_path:
        return None, 0
    print(f"Extracting frame at timestamp: {timestamp}") 
    cap = cv2.VideoCapture(video_path)
    if not cap.isOpened():
        return None, timestamp

    fps = cap.get(cv2.CAP_PROP_FPS)
    if fps <= 0:
        fps = 16.0
    target_frame_num = int(float(timestamp) * fps)
    
    total_frames = int(cap.get(cv2.CAP_PROP_FRAME_COUNT))
    if total_frames > 0 and target_frame_num >= total_frames:
        target_frame_num = total_frames - 1
    
    cap.set(cv2.CAP_PROP_POS_FRAMES, target_frame_num)
    ret, frame = cap.read()
    cap.release()
    
    if ret:
        return cv2.cvtColor(frame, cv2.COLOR_BGR2RGB), timestamp
    return None, timestamp

def resize_image(image: Image.Image) -> Image.Image:
    width, height = image.size
    if width == height:
        return image.resize((SQUARE_DIM, SQUARE_DIM), Image.LANCZOS)
    
    aspect_ratio = width / height
    MAX_ASPECT_RATIO = MAX_DIM / MIN_DIM
    MIN_ASPECT_RATIO = MIN_DIM / MAX_DIM

    image_to_resize = image
    if aspect_ratio > MAX_ASPECT_RATIO:
        target_w, target_h = MAX_DIM, MIN_DIM
        crop_width = int(round(height * MAX_ASPECT_RATIO))
        left = (width - crop_width) // 2
        image_to_resize = image.crop((left, 0, left + crop_width, height))
    elif aspect_ratio < MIN_ASPECT_RATIO:
        target_w, target_h = MIN_DIM, MAX_DIM
        crop_height = int(round(width / MIN_ASPECT_RATIO))
        top = (height - crop_height) // 2
        image_to_resize = image.crop((0, top, width, top + crop_height))
    else:
        if width > height:
            target_w = MAX_DIM
            target_h = int(round(target_w / aspect_ratio))
        else:
            target_h = MAX_DIM
            target_w = int(round(target_h * aspect_ratio))

    final_w = round(target_w / MULTIPLE_OF) * MULTIPLE_OF
    final_h = round(target_h / MULTIPLE_OF) * MULTIPLE_OF
    final_w = max(MIN_DIM, min(MAX_DIM, final_w))
    final_h = max(MIN_DIM, min(MAX_DIM, final_h))
    return image_to_resize.resize((final_w, final_h), Image.LANCZOS)

def resize_and_crop_to_match(target_image, reference_image):
    ref_width, ref_height = reference_image.size
    target_width, target_height = target_image.size
    scale = max(ref_width / target_width, ref_height / target_height)
    new_width, new_height = int(target_width * scale), int(target_height * scale)
    resized = target_image.resize((new_width, new_height), Image.Resampling.LANCZOS)
    left, top = (new_width - ref_width) // 2, (new_height - ref_height) // 2
    return resized.crop((left, top, left + ref_width, top + ref_height))

def get_num_frames(duration_seconds: float):
    raw = int(round(duration_seconds * FIXED_FPS))
    raw = max(MIN_FRAMES_MODEL, min(MAX_FRAMES_MODEL, raw))
    return ((raw - 1) // 4) * 4 + 1