141 lines
4.4 KiB
Python
141 lines
4.4 KiB
Python
import os
|
|
import cv2
|
|
import threading
|
|
import platform
|
|
import torch
|
|
import modules
|
|
import numpy as np
|
|
from typing import Any, List
|
|
from modules.core import update_status
|
|
from modules.face_analyser import get_one_face
|
|
from modules.typing import Frame, Face
|
|
from modules.utilities import (
|
|
conditional_download,
|
|
is_image,
|
|
is_video,
|
|
)
|
|
|
|
FACE_ENHANCER = None
|
|
THREAD_SEMAPHORE = threading.Semaphore()
|
|
THREAD_LOCK = threading.Lock()
|
|
NAME = "DLC.FACE-ENHANCER"
|
|
|
|
abs_dir = os.path.dirname(os.path.abspath(__file__))
|
|
models_dir = os.path.join(
|
|
os.path.dirname(os.path.dirname(os.path.dirname(abs_dir))), "models"
|
|
)
|
|
|
|
|
|
def pre_check() -> bool:
|
|
"""Ensure required model is downloaded."""
|
|
download_directory_path = models_dir
|
|
conditional_download(
|
|
download_directory_path,
|
|
[
|
|
"https://github.com/TencentARC/GFPGAN/releases/download/v1.3.4/GFPGANv1.4.pth"
|
|
],
|
|
)
|
|
return True
|
|
|
|
|
|
def pre_start() -> bool:
|
|
"""Check if target path is valid before starting."""
|
|
if not is_image(modules.globals.target_path) and not is_video(
|
|
modules.globals.target_path
|
|
):
|
|
update_status("Select an image or video for target path.", NAME)
|
|
return False
|
|
return True
|
|
|
|
|
|
TENSORRT_AVAILABLE = False
|
|
try:
|
|
import tensorrt
|
|
TENSORRT_AVAILABLE = True
|
|
except ImportError as im:
|
|
print(f"TensorRT is not available: {im}")
|
|
except Exception as e:
|
|
print(f"TensorRT is not available: {e}")
|
|
|
|
|
|
def get_face_enhancer() -> Any:
|
|
"""Thread-safe singleton loader for the face enhancer model."""
|
|
global FACE_ENHANCER
|
|
with THREAD_LOCK:
|
|
if FACE_ENHANCER is None:
|
|
model_path = os.path.join(models_dir, "GFPGANv1.4.pth")
|
|
selected_device = "cpu"
|
|
if TENSORRT_AVAILABLE and torch.cuda.is_available():
|
|
selected_device = "cuda"
|
|
elif torch.cuda.is_available():
|
|
selected_device = "cuda"
|
|
elif hasattr(torch.backends, "mps") and torch.backends.mps.is_available() and platform.system() == "Darwin":
|
|
selected_device = "mps"
|
|
# Import GFPGAN only when needed
|
|
try:
|
|
import gfpgan
|
|
|
|
FACE_ENHANCER = gfpgan.GFPGANer(model_path=model_path, upscale=1, device=selected_device)
|
|
except Exception as e:
|
|
print(f"Failed to load GFPGAN: {e}")
|
|
FACE_ENHANCER = None
|
|
return FACE_ENHANCER
|
|
|
|
|
|
def enhance_face(temp_frame: Any) -> Any:
|
|
"""Enhance a face in the given frame using GFPGAN."""
|
|
with THREAD_SEMAPHORE:
|
|
enhancer = get_face_enhancer()
|
|
if enhancer is None:
|
|
print("Face enhancer model not loaded.")
|
|
return temp_frame
|
|
try:
|
|
_, _, temp_frame = enhancer.enhance(temp_frame, paste_back=True)
|
|
except Exception as e:
|
|
print(f"Face enhancement failed: {e}")
|
|
return temp_frame
|
|
|
|
|
|
def process_frame(source_face: Any, temp_frame: Any) -> Any:
|
|
"""Process a single frame for face enhancement."""
|
|
target_face = get_one_face(temp_frame)
|
|
if target_face:
|
|
temp_frame = enhance_face(temp_frame)
|
|
return temp_frame
|
|
|
|
|
|
def process_frames(
|
|
source_path: str, temp_frame_paths: List[str], progress: Any = None
|
|
) -> None:
|
|
"""Process a list of frames for face enhancement, updating progress and handling errors."""
|
|
for temp_frame_path in temp_frame_paths:
|
|
temp_frame = cv2.imread(temp_frame_path)
|
|
try:
|
|
result = process_frame(None, temp_frame)
|
|
cv2.imwrite(temp_frame_path, result)
|
|
except Exception as e:
|
|
print(f"Frame enhancement failed: {e}")
|
|
finally:
|
|
if progress:
|
|
progress.update(1)
|
|
|
|
|
|
def process_image(source_path: str, target_path: str, output_path: str) -> None:
|
|
"""Process a single image for face enhancement."""
|
|
target_frame = cv2.imread(target_path)
|
|
result = process_frame(None, target_frame)
|
|
cv2.imwrite(output_path, result)
|
|
|
|
|
|
def process_video(source_path: str, temp_frame_paths: List[str]) -> None:
|
|
"""Process a video for face enhancement."""
|
|
modules.processors.frame.core.process_video(None, temp_frame_paths, process_frames)
|
|
|
|
|
|
def process_frame_v2(temp_frame: Any) -> Any:
|
|
"""Alternative frame processing for face enhancement (for mapped faces, if needed)."""
|
|
target_face = get_one_face(temp_frame)
|
|
if target_face:
|
|
temp_frame = enhance_face(temp_frame)
|
|
return temp_frame
|