Compare commits

..

25 Commits

Author SHA1 Message Date
Kenneth Estanislao 390b88216b Update README.md 2025-02-14 17:33:33 +08:00
Kenneth Estanislao dabaa64695 Merge pull request #932 from harmeetsingh-work/patch-1
Update requirements.txt
2025-02-12 15:21:27 +08:00
Harmeet Singh 1fad1cd43a Update requirements.txt
Made changes for apple silicon. 

Or getting
ERROR: Could not find a version that satisfies the requirement torch==2.5.1+cu118 (from versions: 1.11.0, 1.12.0, 1.12.1, 1.13.0, 1.13.1, 2.0.0, 2.0.1, 2.1.0, 2.1.1, 2.1.2, 2.2.0, 2.2.1, 2.2.2, 2.3.0, 2.3.1, 2.4.0, 2.4.1, 2.5.0, 2.5.1, 2.6.0)
ERROR: No matching distribution found for torch==2.5.1+cu118
2025-02-11 18:44:23 +05:30
Kenneth Estanislao 2f67e2f159 Update requirements.txt 2025-02-09 14:17:49 +08:00
Kenneth Estanislao a3af249ea6 Update requirements.txt 2025-02-07 19:31:02 +08:00
Kenneth Estanislao 5bc3ada632 Update requirements.txt 2025-02-06 15:37:55 +08:00
KRSHH 650e89eb21 Reduced File Size 2025-02-06 10:40:32 +05:30
Kenneth Estanislao 4d2aea37b7 Update requirements.txt 2025-02-06 00:43:20 +08:00
Kenneth Estanislao 28c4b34db1 Merge pull request #911 from nimishgautam/main
Fix cv2 size errors on first run in ui.py
2025-02-05 12:51:39 +08:00
Kenneth Estanislao 49e8f78513 Merge pull request #913 from soulee-dev/main
fix: typo souce_target_map → source_target_map
2025-02-05 12:18:48 +08:00
Kenneth Estanislao d753f5d4b0 Merge pull request #917 from carpusherw/patch-1
Fix requirements.txt
2025-02-05 12:17:42 +08:00
KRSHH 4fb69476d8 Change img dimensions 2025-02-05 12:16:08 +08:00
carpusherw f3adfd194d Fix requirements.txt 2025-02-05 12:16:08 +08:00
Kenneth Estanislao e5f04cf917 Revert "Update requirements.txt"
This reverts commit d45dedc9a6.
2025-02-05 12:08:19 +08:00
Kenneth Estanislao 67394a3157 Revert "Update requirements.txt"
This reverts commit f82cebf86e.
2025-02-05 12:08:10 +08:00
carpusherw 186d155e1b Fix requirements.txt 2025-02-05 09:17:11 +08:00
KRSHH 87081e78d0 Fixed typo 2025-02-04 21:20:54 +05:30
KRSHH f79373d4db Updated Features Section 2025-02-04 21:08:36 +05:30
Soul Lee 513e413956 fix: typo souce_target_map → source_target_map 2025-02-03 20:33:44 +09:00
Kenneth Estanislao f82cebf86e Update requirements.txt 2025-02-03 18:03:27 +08:00
Kenneth Estanislao d45dedc9a6 Update requirements.txt 2025-02-03 16:38:18 +08:00
Kenneth Estanislao 2d489b57ec Update README.md 2025-02-03 13:13:56 +08:00
Nimish Gåtam ccc04983cf Update ui.py
removed unnecessary code as per AI code review (which is a thing now because of course it is)
2025-02-01 12:38:37 +01:00
Nimish Gåtam 2506c5a261 Update ui.py
Some checks for first run when models are missing, so it doesn't error out with inv_scale_x > 0 in cv2
2025-02-01 11:52:49 +01:00
Kenneth Estanislao e862ff1456 Update requirements.txt
updated from CUDA 11.8 to CUDA 12.1
2025-02-01 12:21:55 +08:00
11 changed files with 555 additions and 1117 deletions
-1
View File
@@ -1 +0,0 @@
3.10.0
+10 -3
View File
@@ -49,7 +49,7 @@ Users are expected to use this software responsibly and legally. If using a real
2. Select which camera to use
3. Press live!
## Features & Uses - Everything is real-time
## Features & Uses - Everything is in real-time
### Mouth Mask
@@ -85,7 +85,7 @@ Users are expected to use this software responsibly and legally. If using a real
### Memes
**Create Your most viral meme yet**
**Create Your Most Viral Meme Yet**
<p align="center">
<img src="media/meme.gif" alt="show" width="450">
@@ -93,6 +93,13 @@ Users are expected to use this software responsibly and legally. If using a real
<sub>Created using Many Faces feature in Deep-Live-Cam</sub>
</p>
### Omegle
**Surprise people on Omegle**
<p align="center">
<video src="https://github.com/user-attachments/assets/2e9b9b82-fa04-4b70-9f56-b1f68e7672d0" width="450" controls></video>
</p>
## Installation (Manual)
@@ -146,7 +153,7 @@ brew install python-tk@3.10
**CUDA Execution Provider (Nvidia)**
1. Install [CUDA Toolkit 11.8](https://developer.nvidia.com/cuda-11-8-0-download-archive) or [CUDA Toolkit 12.1.1](https://developer.nvidia.com/cuda-12-1-1-download-archive)
1. Install [CUDA Toolkit 11.8.0](https://developer.nvidia.com/cuda-11-8-0-download-archive)
2. Install dependencies:
```bash
+4 -4
View File
@@ -20,7 +20,6 @@ import modules.metadata
import modules.ui as ui
from modules.processors.frame.core import get_frame_processors_modules
from modules.utilities import has_image_extension, is_image, is_video, detect_fps, create_video, extract_frames, get_temp_frame_paths, restore_audio, create_temp, move_temp, clean_temp, normalize_output_path
from modules.fake_face_handler import cleanup_fake_face
if 'ROCMExecutionProvider' in modules.globals.execution_providers:
del torch
@@ -36,7 +35,9 @@ def parse_args() -> None:
program.add_argument('-t', '--target', help='select an target image or video', dest='target_path')
program.add_argument('-o', '--output', help='select output file or directory', dest='output_path')
program.add_argument('--frame-processor', help='pipeline of frame processors', dest='frame_processor', default=['face_swapper'], choices=['face_swapper', 'face_enhancer'], nargs='+')
program.add_argument('--keep-fps', help='keep original fps', dest='keep_fps', action='store_true', default=False)
program.add_argument('--keep-audio', help='keep original audio', dest='keep_audio', action='store_true', default=True)
program.add_argument('--keep-frames', help='keep temporary frames', dest='keep_frames', action='store_true', default=False)
program.add_argument('--many-faces', help='process every face', dest='many_faces', action='store_true', default=False)
program.add_argument('--nsfw-filter', help='filter the NSFW image or video', dest='nsfw_filter', action='store_true', default=False)
program.add_argument('--map-faces', help='map source target faces', dest='map_faces', action='store_true', default=False)
@@ -64,9 +65,9 @@ def parse_args() -> None:
modules.globals.output_path = normalize_output_path(modules.globals.source_path, modules.globals.target_path, args.output_path)
modules.globals.frame_processors = args.frame_processor
modules.globals.headless = args.source_path or args.target_path or args.output_path
modules.globals.keep_fps = True
modules.globals.keep_frames = True
modules.globals.keep_fps = args.keep_fps
modules.globals.keep_audio = args.keep_audio
modules.globals.keep_frames = args.keep_frames
modules.globals.many_faces = args.many_faces
modules.globals.mouth_mask = args.mouth_mask
modules.globals.nsfw_filter = args.nsfw_filter
@@ -240,7 +241,6 @@ def start() -> None:
def destroy(to_quit=True) -> None:
if modules.globals.target_path:
clean_temp(modules.globals.target_path)
cleanup_fake_face()
if to_quit: quit()
Binary file not shown.

Before

Width:  |  Height:  |  Size: 264 KiB

+12 -12
View File
@@ -39,13 +39,13 @@ def get_many_faces(frame: Frame) -> Any:
return None
def has_valid_map() -> bool:
for map in modules.globals.souce_target_map:
for map in modules.globals.source_target_map:
if "source" in map and "target" in map:
return True
return False
def default_source_face() -> Any:
for map in modules.globals.souce_target_map:
for map in modules.globals.source_target_map:
if "source" in map:
return map['source']['face']
return None
@@ -53,7 +53,7 @@ def default_source_face() -> Any:
def simplify_maps() -> Any:
centroids = []
faces = []
for map in modules.globals.souce_target_map:
for map in modules.globals.source_target_map:
if "source" in map and "target" in map:
centroids.append(map['target']['face'].normed_embedding)
faces.append(map['source']['face'])
@@ -64,10 +64,10 @@ def simplify_maps() -> Any:
def add_blank_map() -> Any:
try:
max_id = -1
if len(modules.globals.souce_target_map) > 0:
max_id = max(modules.globals.souce_target_map, key=lambda x: x['id'])['id']
if len(modules.globals.source_target_map) > 0:
max_id = max(modules.globals.source_target_map, key=lambda x: x['id'])['id']
modules.globals.souce_target_map.append({
modules.globals.source_target_map.append({
'id' : max_id + 1
})
except ValueError:
@@ -75,14 +75,14 @@ def add_blank_map() -> Any:
def get_unique_faces_from_target_image() -> Any:
try:
modules.globals.souce_target_map = []
modules.globals.source_target_map = []
target_frame = cv2.imread(modules.globals.target_path)
many_faces = get_many_faces(target_frame)
i = 0
for face in many_faces:
x_min, y_min, x_max, y_max = face['bbox']
modules.globals.souce_target_map.append({
modules.globals.source_target_map.append({
'id' : i,
'target' : {
'cv2' : target_frame[int(y_min):int(y_max), int(x_min):int(x_max)],
@@ -96,7 +96,7 @@ def get_unique_faces_from_target_image() -> Any:
def get_unique_faces_from_target_video() -> Any:
try:
modules.globals.souce_target_map = []
modules.globals.source_target_map = []
frame_face_embeddings = []
face_embeddings = []
@@ -127,7 +127,7 @@ def get_unique_faces_from_target_video() -> Any:
face['target_centroid'] = closest_centroid_index
for i in range(len(centroids)):
modules.globals.souce_target_map.append({
modules.globals.source_target_map.append({
'id' : i
})
@@ -135,7 +135,7 @@ def get_unique_faces_from_target_video() -> Any:
for frame in tqdm(frame_face_embeddings, desc=f"Mapping frame embeddings to centroids-{i}"):
temp.append({'frame': frame['frame'], 'faces': [face for face in frame['faces'] if face['target_centroid'] == i], 'location': frame['location']})
modules.globals.souce_target_map[i]['target_faces_in_frame'] = temp
modules.globals.source_target_map[i]['target_faces_in_frame'] = temp
# dump_faces(centroids, frame_face_embeddings)
default_target_face()
@@ -144,7 +144,7 @@ def get_unique_faces_from_target_video() -> Any:
def default_target_face():
for map in modules.globals.souce_target_map:
for map in modules.globals.source_target_map:
best_face = None
best_frame = None
for frame in map['target_faces_in_frame']:
-120
View File
@@ -1,120 +0,0 @@
import os
import requests
import tempfile
from pathlib import Path
import cv2
import numpy as np
import modules.globals
def add_padding_to_face(image, padding_ratio=0.3):
"""Add padding around the face image
Args:
image: The input face image
padding_ratio: Amount of padding to add as a ratio of image dimensions
Returns:
Padded image with background padding added
"""
if image is None:
return None
height, width = image.shape[:2]
pad_x = int(width * padding_ratio)
pad_y = int(height * padding_ratio)
# Create larger image with padding
padded_height = height + 2 * pad_y
padded_width = width + 2 * pad_x
padded_image = np.zeros((padded_height, padded_width, 3), dtype=np.uint8)
# Fill padded area with blurred and darkened edge pixels
edge_color = cv2.blur(image, (15, 15))
edge_color = (edge_color * 0.6).astype(np.uint8) # Darken the padding
# Fill the padded image with original face
padded_image[pad_y:pad_y+height, pad_x:pad_x+width] = image
# Fill padding areas with edge color
# Top padding - repeat first row
top_edge = edge_color[0, :, :]
for i in range(pad_y):
padded_image[i, pad_x:pad_x+width] = top_edge
# Bottom padding - repeat last row
bottom_edge = edge_color[-1, :, :]
for i in range(pad_y):
padded_image[pad_y+height+i, pad_x:pad_x+width] = bottom_edge
# Left padding - repeat first column
left_edge = edge_color[:, 0, :]
for i in range(pad_x):
padded_image[pad_y:pad_y+height, i] = left_edge
# Right padding - repeat last column
right_edge = edge_color[:, -1, :]
for i in range(pad_x):
padded_image[pad_y:pad_y+height, pad_x+width+i] = right_edge
# Fill corners with nearest edge colors
# Top-left corner
padded_image[:pad_y, :pad_x] = edge_color[0, 0, :]
# Top-right corner
padded_image[:pad_y, pad_x+width:] = edge_color[0, -1, :]
# Bottom-left corner
padded_image[pad_y+height:, :pad_x] = edge_color[-1, 0, :]
# Bottom-right corner
padded_image[pad_y+height:, pad_x+width:] = edge_color[-1, -1, :]
return padded_image
def get_fake_face() -> str:
"""Fetch a face from thispersondoesnotexist.com and save it temporarily"""
try:
# Create temp directory if it doesn't exist
temp_dir = Path(tempfile.gettempdir()) / "deep-live-cam"
temp_dir.mkdir(parents=True, exist_ok=True)
# Generate temp file path
temp_file = temp_dir / "fake_face.jpg"
# Basic headers to mimic a browser request
headers = {
'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/91.0.4472.124 Safari/537.36'
}
# Fetch the image
response = requests.get('https://thispersondoesnotexist.com', headers=headers)
if response.status_code == 200:
# Read image from response
image_array = np.asarray(bytearray(response.content), dtype=np.uint8)
image = cv2.imdecode(image_array, cv2.IMREAD_COLOR)
# Add padding around the face
padded_image = add_padding_to_face(image)
# Save the padded image
cv2.imwrite(str(temp_file), padded_image)
return str(temp_file)
else:
print(f"Failed to fetch fake face: {response.status_code}")
return None
except Exception as e:
print(f"Error fetching fake face: {str(e)}")
return None
def cleanup_fake_face():
"""Clean up the temporary fake face image"""
try:
if modules.globals.fake_face_path and os.path.exists(modules.globals.fake_face_path):
os.remove(modules.globals.fake_face_path)
modules.globals.fake_face_path = None
except Exception as e:
print(f"Error cleaning up fake face: {str(e)}")
def refresh_fake_face():
"""Refresh the fake face image"""
cleanup_fake_face()
modules.globals.fake_face_path = get_fake_face()
return modules.globals.fake_face_path is not None
+2 -11
View File
@@ -9,7 +9,7 @@ file_types = [
("Video", ("*.mp4", "*.mkv")),
]
souce_target_map = []
source_target_map = []
simple_map = {}
source_path = None
@@ -21,7 +21,7 @@ keep_audio = True
keep_frames = False
many_faces = False
map_faces = False
color_correction = False
color_correction = False # New global variable for color correction toggle
nsfw_filter = False
video_encoder = None
video_quality = None
@@ -41,12 +41,3 @@ show_mouth_mask_box = False
mask_feather_ratio = 8
mask_down_size = 0.50
mask_size = 1
mouth_mask_size = 1.0
eyes_mask = False
show_eyes_mask_box = False
eyebrows_mask = False
show_eyebrows_mask_box = False
eyes_mask_size = 1.0
eyebrows_mask_size = 1.0
use_fake_face = False
fake_face_path = None
-634
View File
@@ -1,634 +0,0 @@
import cv2
import numpy as np
from modules.typing import Face, Frame
import modules.globals
def apply_color_transfer(source, target):
"""
Apply color transfer from target to source image
"""
source = cv2.cvtColor(source, cv2.COLOR_BGR2LAB).astype("float32")
target = cv2.cvtColor(target, cv2.COLOR_BGR2LAB).astype("float32")
source_mean, source_std = cv2.meanStdDev(source)
target_mean, target_std = cv2.meanStdDev(target)
# Reshape mean and std to be broadcastable
source_mean = source_mean.reshape(1, 1, 3)
source_std = source_std.reshape(1, 1, 3)
target_mean = target_mean.reshape(1, 1, 3)
target_std = target_std.reshape(1, 1, 3)
# Perform the color transfer
source = (source - source_mean) * (target_std / source_std) + target_mean
return cv2.cvtColor(np.clip(source, 0, 255).astype("uint8"), cv2.COLOR_LAB2BGR)
def create_face_mask(face: Face, frame: Frame) -> np.ndarray:
mask = np.zeros(frame.shape[:2], dtype=np.uint8)
landmarks = face.landmark_2d_106
if landmarks is not None:
# Convert landmarks to int32
landmarks = landmarks.astype(np.int32)
# Extract facial features
right_side_face = landmarks[0:16]
left_side_face = landmarks[17:32]
right_eye = landmarks[33:42]
right_eye_brow = landmarks[43:51]
left_eye = landmarks[87:96]
left_eye_brow = landmarks[97:105]
# Calculate forehead extension
right_eyebrow_top = np.min(right_eye_brow[:, 1])
left_eyebrow_top = np.min(left_eye_brow[:, 1])
eyebrow_top = min(right_eyebrow_top, left_eyebrow_top)
face_top = np.min([right_side_face[0, 1], left_side_face[-1, 1]])
forehead_height = face_top - eyebrow_top
extended_forehead_height = int(forehead_height * 5.0) # Extend by 50%
# Create forehead points
forehead_left = right_side_face[0].copy()
forehead_right = left_side_face[-1].copy()
forehead_left[1] -= extended_forehead_height
forehead_right[1] -= extended_forehead_height
# Combine all points to create the face outline
face_outline = np.vstack(
[
[forehead_left],
right_side_face,
left_side_face[::-1], # Reverse left side to create a continuous outline
[forehead_right],
]
)
# Calculate padding
padding = int(
np.linalg.norm(right_side_face[0] - left_side_face[-1]) * 0.05
) # 5% of face width
# Create a slightly larger convex hull for padding
hull = cv2.convexHull(face_outline)
hull_padded = []
for point in hull:
x, y = point[0]
center = np.mean(face_outline, axis=0)
direction = np.array([x, y]) - center
direction = direction / np.linalg.norm(direction)
padded_point = np.array([x, y]) + direction * padding
hull_padded.append(padded_point)
hull_padded = np.array(hull_padded, dtype=np.int32)
# Fill the padded convex hull
cv2.fillConvexPoly(mask, hull_padded, 255)
# Smooth the mask edges
mask = cv2.GaussianBlur(mask, (5, 5), 3)
return mask
def create_lower_mouth_mask(
face: Face, frame: Frame
) -> (np.ndarray, np.ndarray, tuple, np.ndarray):
mask = np.zeros(frame.shape[:2], dtype=np.uint8)
mouth_cutout = None
landmarks = face.landmark_2d_106
if landmarks is not None:
# 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
lower_lip_order = [
65,
66,
62,
70,
69,
18,
19,
20,
21,
22,
23,
24,
0,
8,
7,
6,
5,
4,
3,
2,
65,
]
lower_lip_landmarks = landmarks[lower_lip_order].astype(
np.float32
) # Use float for precise calculations
# Calculate the center of the landmarks
center = np.mean(lower_lip_landmarks, axis=0)
# Expand the landmarks outward using the mouth_mask_size
expansion_factor = (
1 + modules.globals.mask_down_size * modules.globals.mouth_mask_size
) # Adjust expansion based on slider
expanded_landmarks = (lower_lip_landmarks - center) * expansion_factor + center
# Extend the top lip part
toplip_indices = [
20,
0,
1,
2,
3,
4,
5,
] # Indices for landmarks 2, 65, 66, 62, 70, 69, 18
toplip_extension = (
modules.globals.mask_size * modules.globals.mouth_mask_size * 0.5
) # Adjust extension based on slider
for idx in toplip_indices:
direction = expanded_landmarks[idx] - center
direction = direction / np.linalg.norm(direction)
expanded_landmarks[idx] += direction * toplip_extension
# Extend the bottom part (chin area)
chin_indices = [
11,
12,
13,
14,
15,
16,
] # Indices for landmarks 21, 22, 23, 24, 0, 8
chin_extension = 2 * 0.2 # Adjust this factor to control the extension
for idx in chin_indices:
expanded_landmarks[idx][1] += (
expanded_landmarks[idx][1] - center[1]
) * chin_extension
# Convert back to integer coordinates
expanded_landmarks = expanded_landmarks.astype(np.int32)
# Calculate bounding box for the expanded lower mouth
min_x, min_y = np.min(expanded_landmarks, axis=0)
max_x, max_y = np.max(expanded_landmarks, axis=0)
# Add some padding to the bounding box
padding = int((max_x - min_x) * 0.1) # 10% padding
min_x = max(0, min_x - padding)
min_y = max(0, min_y - padding)
max_x = min(frame.shape[1], max_x + padding)
max_y = min(frame.shape[0], max_y + padding)
# Ensure the bounding box dimensions are valid
if max_x <= min_x or max_y <= min_y:
if (max_x - min_x) <= 1:
max_x = min_x + 1
if (max_y - min_y) <= 1:
max_y = min_y + 1
# Create the mask
mask_roi = np.zeros((max_y - min_y, max_x - min_x), dtype=np.uint8)
cv2.fillPoly(mask_roi, [expanded_landmarks - [min_x, min_y]], 255)
# Apply Gaussian blur to soften the mask edges
mask_roi = cv2.GaussianBlur(mask_roi, (15, 15), 5)
# Place the mask ROI in the full-sized mask
mask[min_y:max_y, min_x:max_x] = mask_roi
# Extract the masked area from the frame
mouth_cutout = frame[min_y:max_y, min_x:max_x].copy()
# Return the expanded lower lip polygon in original frame coordinates
lower_lip_polygon = expanded_landmarks
return mask, mouth_cutout, (min_x, min_y, max_x, max_y), lower_lip_polygon
def create_eyes_mask(face: Face, frame: Frame) -> (np.ndarray, np.ndarray, tuple, np.ndarray):
mask = np.zeros(frame.shape[:2], dtype=np.uint8)
eyes_cutout = None
landmarks = face.landmark_2d_106
if landmarks is not None:
# Left eye landmarks (87-96) and right eye landmarks (33-42)
left_eye = landmarks[87:96]
right_eye = landmarks[33:42]
# Calculate centers and dimensions for each eye
left_eye_center = np.mean(left_eye, axis=0).astype(np.int32)
right_eye_center = np.mean(right_eye, axis=0).astype(np.int32)
# Calculate eye dimensions with size adjustment
def get_eye_dimensions(eye_points):
x_coords = eye_points[:, 0]
y_coords = eye_points[:, 1]
width = int((np.max(x_coords) - np.min(x_coords)) * (1 + modules.globals.mask_down_size * modules.globals.eyes_mask_size))
height = int((np.max(y_coords) - np.min(y_coords)) * (1 + modules.globals.mask_down_size * modules.globals.eyes_mask_size))
return width, height
left_width, left_height = get_eye_dimensions(left_eye)
right_width, right_height = get_eye_dimensions(right_eye)
# Add extra padding
padding = int(max(left_width, right_width) * 0.2)
# Calculate bounding box for both eyes
min_x = min(left_eye_center[0] - left_width//2, right_eye_center[0] - right_width//2) - padding
max_x = max(left_eye_center[0] + left_width//2, right_eye_center[0] + right_width//2) + padding
min_y = min(left_eye_center[1] - left_height//2, right_eye_center[1] - right_height//2) - padding
max_y = max(left_eye_center[1] + left_height//2, right_eye_center[1] + right_height//2) + padding
# Ensure coordinates are within frame bounds
min_x = max(0, min_x)
min_y = max(0, min_y)
max_x = min(frame.shape[1], max_x)
max_y = min(frame.shape[0], max_y)
# Create mask for the eyes region
mask_roi = np.zeros((max_y - min_y, max_x - min_x), dtype=np.uint8)
# Draw ellipses for both eyes
left_center = (left_eye_center[0] - min_x, left_eye_center[1] - min_y)
right_center = (right_eye_center[0] - min_x, right_eye_center[1] - min_y)
# Calculate axes lengths (half of width and height)
left_axes = (left_width//2, left_height//2)
right_axes = (right_width//2, right_height//2)
# Draw filled ellipses
cv2.ellipse(mask_roi, left_center, left_axes, 0, 0, 360, 255, -1)
cv2.ellipse(mask_roi, right_center, right_axes, 0, 0, 360, 255, -1)
# Apply Gaussian blur to soften mask edges
mask_roi = cv2.GaussianBlur(mask_roi, (15, 15), 5)
# Place the mask ROI in the full-sized mask
mask[min_y:max_y, min_x:max_x] = mask_roi
# Extract the masked area from the frame
eyes_cutout = frame[min_y:max_y, min_x:max_x].copy()
# Create polygon points for visualization
def create_ellipse_points(center, axes):
t = np.linspace(0, 2*np.pi, 32)
x = center[0] + axes[0] * np.cos(t)
y = center[1] + axes[1] * np.sin(t)
return np.column_stack((x, y)).astype(np.int32)
# Generate points for both ellipses
left_points = create_ellipse_points((left_eye_center[0], left_eye_center[1]), (left_width//2, left_height//2))
right_points = create_ellipse_points((right_eye_center[0], right_eye_center[1]), (right_width//2, right_height//2))
# Combine points for both eyes
eyes_polygon = np.vstack([left_points, right_points])
return mask, eyes_cutout, (min_x, min_y, max_x, max_y), eyes_polygon
def create_curved_eyebrow(points):
if len(points) >= 5:
# Sort points by x-coordinate
sorted_idx = np.argsort(points[:, 0])
sorted_points = points[sorted_idx]
# Calculate dimensions
x_min, y_min = np.min(sorted_points, axis=0)
x_max, y_max = np.max(sorted_points, axis=0)
width = x_max - x_min
height = y_max - y_min
# Create more points for smoother curve
num_points = 50
x = np.linspace(x_min, x_max, num_points)
# Fit quadratic curve through points for more natural arch
coeffs = np.polyfit(sorted_points[:, 0], sorted_points[:, 1], 2)
y = np.polyval(coeffs, x)
# Increased offsets to create more separation
top_offset = height * 0.5 # Increased from 0.3 to shift up more
bottom_offset = height * 0.2 # Increased from 0.1 to shift down more
# Create smooth curves
top_curve = y - top_offset
bottom_curve = y + bottom_offset
# Create curved endpoints with more pronounced taper
end_points = 5
start_x = np.linspace(x[0] - width * 0.15, x[0], end_points) # Increased taper
end_x = np.linspace(x[-1], x[-1] + width * 0.15, end_points) # Increased taper
# Create tapered ends
start_curve = np.column_stack((
start_x,
np.linspace(bottom_curve[0], top_curve[0], end_points)
))
end_curve = np.column_stack((
end_x,
np.linspace(bottom_curve[-1], top_curve[-1], end_points)
))
# Combine all points to form a smooth contour
contour_points = np.vstack([
start_curve,
np.column_stack((x, top_curve)),
end_curve,
np.column_stack((x[::-1], bottom_curve[::-1]))
])
# Add slight padding for better coverage
center = np.mean(contour_points, axis=0)
vectors = contour_points - center
padded_points = center + vectors * 1.2 # Increased padding slightly
return padded_points
return points
def create_eyebrows_mask(face: Face, frame: Frame) -> (np.ndarray, np.ndarray, tuple, np.ndarray):
mask = np.zeros(frame.shape[:2], dtype=np.uint8)
eyebrows_cutout = None
landmarks = face.landmark_2d_106
if landmarks is not None:
# Left eyebrow landmarks (97-105) and right eyebrow landmarks (43-51)
left_eyebrow = landmarks[97:105].astype(np.float32)
right_eyebrow = landmarks[43:51].astype(np.float32)
# Calculate centers and dimensions for each eyebrow
left_center = np.mean(left_eyebrow, axis=0)
right_center = np.mean(right_eyebrow, axis=0)
# Calculate bounding box with padding adjusted by size
all_points = np.vstack([left_eyebrow, right_eyebrow])
padding_factor = modules.globals.eyebrows_mask_size
min_x = np.min(all_points[:, 0]) - 25 * padding_factor
max_x = np.max(all_points[:, 0]) + 25 * padding_factor
min_y = np.min(all_points[:, 1]) - 20 * padding_factor
max_y = np.max(all_points[:, 1]) + 15 * padding_factor
# Ensure coordinates are within frame bounds
min_x = max(0, int(min_x))
min_y = max(0, int(min_y))
max_x = min(frame.shape[1], int(max_x))
max_y = min(frame.shape[0], int(max_y))
# Create mask for the eyebrows region
mask_roi = np.zeros((max_y - min_y, max_x - min_x), dtype=np.uint8)
try:
# Convert points to local coordinates
left_local = left_eyebrow - [min_x, min_y]
right_local = right_eyebrow - [min_x, min_y]
def create_curved_eyebrow(points):
if len(points) >= 5:
# Sort points by x-coordinate
sorted_idx = np.argsort(points[:, 0])
sorted_points = points[sorted_idx]
# Calculate dimensions
x_min, y_min = np.min(sorted_points, axis=0)
x_max, y_max = np.max(sorted_points, axis=0)
width = x_max - x_min
height = y_max - y_min
# Create more points for smoother curve
num_points = 50
x = np.linspace(x_min, x_max, num_points)
# Fit quadratic curve through points for more natural arch
coeffs = np.polyfit(sorted_points[:, 0], sorted_points[:, 1], 2)
y = np.polyval(coeffs, x)
# Increased offsets to create more separation
top_offset = height * 0.5 # Increased from 0.3 to shift up more
bottom_offset = height * 0.2 # Increased from 0.1 to shift down more
# Create smooth curves
top_curve = y - top_offset
bottom_curve = y + bottom_offset
# Create curved endpoints with more pronounced taper
end_points = 5
start_x = np.linspace(x[0] - width * 0.15, x[0], end_points) # Increased taper
end_x = np.linspace(x[-1], x[-1] + width * 0.15, end_points) # Increased taper
# Create tapered ends
start_curve = np.column_stack((
start_x,
np.linspace(bottom_curve[0], top_curve[0], end_points)
))
end_curve = np.column_stack((
end_x,
np.linspace(bottom_curve[-1], top_curve[-1], end_points)
))
# Combine all points to form a smooth contour
contour_points = np.vstack([
start_curve,
np.column_stack((x, top_curve)),
end_curve,
np.column_stack((x[::-1], bottom_curve[::-1]))
])
# Add slight padding for better coverage
center = np.mean(contour_points, axis=0)
vectors = contour_points - center
padded_points = center + vectors * 1.2 # Increased padding slightly
return padded_points
return points
# Generate and draw eyebrow shapes
left_shape = create_curved_eyebrow(left_local)
right_shape = create_curved_eyebrow(right_local)
# Apply multi-stage blurring for natural feathering
# First, strong Gaussian blur for initial softening
mask_roi = cv2.GaussianBlur(mask_roi, (21, 21), 7)
# Second, medium blur for transition areas
mask_roi = cv2.GaussianBlur(mask_roi, (11, 11), 3)
# Finally, light blur for fine details
mask_roi = cv2.GaussianBlur(mask_roi, (5, 5), 1)
# Normalize mask values
mask_roi = cv2.normalize(mask_roi, None, 0, 255, cv2.NORM_MINMAX)
# Place the mask ROI in the full-sized mask
mask[min_y:max_y, min_x:max_x] = mask_roi
# Extract the masked area from the frame
eyebrows_cutout = frame[min_y:max_y, min_x:max_x].copy()
# Combine points for visualization
eyebrows_polygon = np.vstack([
left_shape + [min_x, min_y],
right_shape + [min_x, min_y]
]).astype(np.int32)
except Exception as e:
# Fallback to simple polygons if curve fitting fails
left_local = left_eyebrow - [min_x, min_y]
right_local = right_eyebrow - [min_x, min_y]
cv2.fillPoly(mask_roi, [left_local.astype(np.int32)], 255)
cv2.fillPoly(mask_roi, [right_local.astype(np.int32)], 255)
mask_roi = cv2.GaussianBlur(mask_roi, (21, 21), 7)
mask[min_y:max_y, min_x:max_x] = mask_roi
eyebrows_cutout = frame[min_y:max_y, min_x:max_x].copy()
eyebrows_polygon = np.vstack([left_eyebrow, right_eyebrow]).astype(np.int32)
return mask, eyebrows_cutout, (min_x, min_y, max_x, max_y), eyebrows_polygon
def apply_mask_area(
frame: np.ndarray,
cutout: np.ndarray,
box: tuple,
face_mask: np.ndarray,
polygon: np.ndarray,
) -> np.ndarray:
min_x, min_y, max_x, max_y = box
box_width = max_x - min_x
box_height = max_y - min_y
if (
cutout is None
or box_width is None
or box_height is None
or face_mask is None
or polygon is None
):
return frame
try:
resized_cutout = cv2.resize(cutout, (box_width, box_height))
roi = frame[min_y:max_y, min_x:max_x]
if roi.shape != resized_cutout.shape:
resized_cutout = cv2.resize(
resized_cutout, (roi.shape[1], roi.shape[0])
)
color_corrected_area = apply_color_transfer(resized_cutout, roi)
# Create mask for the area
polygon_mask = np.zeros(roi.shape[:2], dtype=np.uint8)
# Split points for left and right parts if needed
if len(polygon) > 50: # Arbitrary threshold to detect if we have multiple parts
mid_point = len(polygon) // 2
left_points = polygon[:mid_point] - [min_x, min_y]
right_points = polygon[mid_point:] - [min_x, min_y]
cv2.fillPoly(polygon_mask, [left_points], 255)
cv2.fillPoly(polygon_mask, [right_points], 255)
else:
adjusted_polygon = polygon - [min_x, min_y]
cv2.fillPoly(polygon_mask, [adjusted_polygon], 255)
# Apply strong initial feathering
polygon_mask = cv2.GaussianBlur(polygon_mask, (21, 21), 7)
# Apply additional feathering
feather_amount = min(
30,
box_width // modules.globals.mask_feather_ratio,
box_height // modules.globals.mask_feather_ratio,
)
feathered_mask = cv2.GaussianBlur(
polygon_mask.astype(float), (0, 0), feather_amount
)
feathered_mask = feathered_mask / feathered_mask.max()
# Apply additional smoothing to the mask edges
feathered_mask = cv2.GaussianBlur(feathered_mask, (5, 5), 1)
face_mask_roi = face_mask[min_y:max_y, min_x:max_x]
combined_mask = feathered_mask * (face_mask_roi / 255.0)
combined_mask = combined_mask[:, :, np.newaxis]
blended = (
color_corrected_area * combined_mask + roi * (1 - combined_mask)
).astype(np.uint8)
# Apply face mask to blended result
face_mask_3channel = (
np.repeat(face_mask_roi[:, :, np.newaxis], 3, axis=2) / 255.0
)
final_blend = blended * face_mask_3channel + roi * (1 - face_mask_3channel)
frame[min_y:max_y, min_x:max_x] = final_blend.astype(np.uint8)
except Exception as e:
pass
return frame
def draw_mask_visualization(
frame: Frame,
mask_data: tuple,
label: str,
draw_method: str = "polygon"
) -> Frame:
mask, cutout, (min_x, min_y, max_x, max_y), polygon = mask_data
vis_frame = frame.copy()
# Ensure coordinates are within frame bounds
height, width = vis_frame.shape[:2]
min_x, min_y = max(0, min_x), max(0, min_y)
max_x, max_y = min(width, max_x), min(height, max_y)
if draw_method == "ellipse" and len(polygon) > 50: # For eyes
# Split points for left and right parts
mid_point = len(polygon) // 2
left_points = polygon[:mid_point]
right_points = polygon[mid_point:]
try:
# Fit ellipses to points - need at least 5 points
if len(left_points) >= 5 and len(right_points) >= 5:
# Convert points to the correct format for ellipse fitting
left_points = left_points.astype(np.float32)
right_points = right_points.astype(np.float32)
# Fit ellipses
left_ellipse = cv2.fitEllipse(left_points)
right_ellipse = cv2.fitEllipse(right_points)
# Draw the ellipses
cv2.ellipse(vis_frame, left_ellipse, (0, 255, 0), 2)
cv2.ellipse(vis_frame, right_ellipse, (0, 255, 0), 2)
except Exception as e:
# If ellipse fitting fails, draw simple rectangles as fallback
left_rect = cv2.boundingRect(left_points)
right_rect = cv2.boundingRect(right_points)
cv2.rectangle(vis_frame,
(left_rect[0], left_rect[1]),
(left_rect[0] + left_rect[2], left_rect[1] + left_rect[3]),
(0, 255, 0), 2)
cv2.rectangle(vis_frame,
(right_rect[0], right_rect[1]),
(right_rect[0] + right_rect[2], right_rect[1] + right_rect[3]),
(0, 255, 0), 2)
else: # For mouth and eyebrows
# Draw the polygon
if len(polygon) > 50: # If we have multiple parts
mid_point = len(polygon) // 2
left_points = polygon[:mid_point]
right_points = polygon[mid_point:]
cv2.polylines(vis_frame, [left_points], True, (0, 255, 0), 2, cv2.LINE_AA)
cv2.polylines(vis_frame, [right_points], True, (0, 255, 0), 2, cv2.LINE_AA)
else:
cv2.polylines(vis_frame, [polygon], True, (0, 255, 0), 2, cv2.LINE_AA)
# Add label
cv2.putText(
vis_frame,
label,
(min_x, min_y - 10),
cv2.FONT_HERSHEY_SIMPLEX,
0.5,
(255, 255, 255),
1,
)
return vis_frame
+376 -64
View File
@@ -14,14 +14,6 @@ from modules.utilities import (
is_video,
)
from modules.cluster_analysis import find_closest_centroid
from modules.processors.frame.face_masking import (
create_face_mask,
create_lower_mouth_mask,
create_eyes_mask,
create_eyebrows_mask,
apply_mask_area,
draw_mask_visualization
)
import os
FACE_SWAPPER = None
@@ -82,62 +74,24 @@ def swap_face(source_face: Face, target_face: Face, temp_frame: Frame) -> Frame:
temp_frame, target_face, source_face, paste_back=True
)
# Create face mask for both mouth and eyes masking
face_mask = create_face_mask(target_face, temp_frame)
if modules.globals.mouth_mask:
# Create and apply mouth mask
mouth_mask_data = create_lower_mouth_mask(target_face, temp_frame)
swapped_frame = apply_mask_area(
swapped_frame,
mouth_mask_data[1], # mouth_cutout
mouth_mask_data[2], # mouth_box
face_mask,
mouth_mask_data[3] # mouth_polygon
# Create a mask for the target face
face_mask = create_face_mask(target_face, temp_frame)
# Create the mouth mask
mouth_mask, mouth_cutout, mouth_box, lower_lip_polygon = (
create_lower_mouth_mask(target_face, temp_frame)
)
# Apply the mouth area
swapped_frame = apply_mouth_area(
swapped_frame, mouth_cutout, mouth_box, face_mask, lower_lip_polygon
)
if modules.globals.show_mouth_mask_box:
swapped_frame = draw_mask_visualization(
swapped_frame,
mouth_mask_data,
"Lower Mouth Mask"
)
if modules.globals.eyes_mask:
# Create and apply eyes mask
eyes_mask_data = create_eyes_mask(target_face, temp_frame)
swapped_frame = apply_mask_area(
swapped_frame,
eyes_mask_data[1], # eyes_cutout
eyes_mask_data[2], # eyes_box
face_mask,
eyes_mask_data[3] # eyes_polygon
)
if modules.globals.show_eyes_mask_box:
swapped_frame = draw_mask_visualization(
swapped_frame,
eyes_mask_data,
"Eyes Mask",
draw_method="ellipse"
)
if modules.globals.eyebrows_mask:
# Create and apply eyebrows mask
eyebrows_mask_data = create_eyebrows_mask(target_face, temp_frame)
swapped_frame = apply_mask_area(
swapped_frame,
eyebrows_mask_data[1], # eyebrows_cutout
eyebrows_mask_data[2], # eyebrows_box
face_mask,
eyebrows_mask_data[3] # eyebrows_polygon
)
if modules.globals.show_eyebrows_mask_box:
swapped_frame = draw_mask_visualization(
swapped_frame,
eyebrows_mask_data,
"Eyebrows Mask"
mouth_mask_data = (mouth_mask, mouth_cutout, mouth_box, lower_lip_polygon)
swapped_frame = draw_mouth_mask_visualization(
swapped_frame, target_face, mouth_mask_data
)
return swapped_frame
@@ -163,12 +117,12 @@ def process_frame_v2(temp_frame: Frame, temp_frame_path: str = "") -> Frame:
if is_image(modules.globals.target_path):
if modules.globals.many_faces:
source_face = default_source_face()
for map in modules.globals.souce_target_map:
for map in modules.globals.source_target_map:
target_face = map["target"]["face"]
temp_frame = swap_face(source_face, target_face, temp_frame)
elif not modules.globals.many_faces:
for map in modules.globals.souce_target_map:
for map in modules.globals.source_target_map:
if "source" in map:
source_face = map["source"]["face"]
target_face = map["target"]["face"]
@@ -177,7 +131,7 @@ def process_frame_v2(temp_frame: Frame, temp_frame_path: str = "") -> Frame:
elif is_video(modules.globals.target_path):
if modules.globals.many_faces:
source_face = default_source_face()
for map in modules.globals.souce_target_map:
for map in modules.globals.source_target_map:
target_frame = [
f
for f in map["target_faces_in_frame"]
@@ -189,7 +143,7 @@ def process_frame_v2(temp_frame: Frame, temp_frame_path: str = "") -> Frame:
temp_frame = swap_face(source_face, target_face, temp_frame)
elif not modules.globals.many_faces:
for map in modules.globals.souce_target_map:
for map in modules.globals.source_target_map:
if "source" in map:
target_frame = [
f
@@ -301,3 +255,361 @@ def process_video(source_path: str, temp_frame_paths: List[str]) -> None:
modules.processors.frame.core.process_video(
source_path, temp_frame_paths, process_frames
)
def create_lower_mouth_mask(
face: Face, frame: Frame
) -> (np.ndarray, np.ndarray, tuple, np.ndarray):
mask = np.zeros(frame.shape[:2], dtype=np.uint8)
mouth_cutout = None
landmarks = face.landmark_2d_106
if landmarks is not None:
# 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
lower_lip_order = [
65,
66,
62,
70,
69,
18,
19,
20,
21,
22,
23,
24,
0,
8,
7,
6,
5,
4,
3,
2,
65,
]
lower_lip_landmarks = landmarks[lower_lip_order].astype(
np.float32
) # Use float for precise calculations
# Calculate the center of the landmarks
center = np.mean(lower_lip_landmarks, axis=0)
# Expand the landmarks outward
expansion_factor = (
1 + modules.globals.mask_down_size
) # Adjust this for more or less expansion
expanded_landmarks = (lower_lip_landmarks - center) * expansion_factor + center
# Extend the top lip part
toplip_indices = [
20,
0,
1,
2,
3,
4,
5,
] # Indices for landmarks 2, 65, 66, 62, 70, 69, 18
toplip_extension = (
modules.globals.mask_size * 0.5
) # Adjust this factor to control the extension
for idx in toplip_indices:
direction = expanded_landmarks[idx] - center
direction = direction / np.linalg.norm(direction)
expanded_landmarks[idx] += direction * toplip_extension
# Extend the bottom part (chin area)
chin_indices = [
11,
12,
13,
14,
15,
16,
] # Indices for landmarks 21, 22, 23, 24, 0, 8
chin_extension = 2 * 0.2 # Adjust this factor to control the extension
for idx in chin_indices:
expanded_landmarks[idx][1] += (
expanded_landmarks[idx][1] - center[1]
) * chin_extension
# Convert back to integer coordinates
expanded_landmarks = expanded_landmarks.astype(np.int32)
# Calculate bounding box for the expanded lower mouth
min_x, min_y = np.min(expanded_landmarks, axis=0)
max_x, max_y = np.max(expanded_landmarks, axis=0)
# Add some padding to the bounding box
padding = int((max_x - min_x) * 0.1) # 10% padding
min_x = max(0, min_x - padding)
min_y = max(0, min_y - padding)
max_x = min(frame.shape[1], max_x + padding)
max_y = min(frame.shape[0], max_y + padding)
# Ensure the bounding box dimensions are valid
if max_x <= min_x or max_y <= min_y:
if (max_x - min_x) <= 1:
max_x = min_x + 1
if (max_y - min_y) <= 1:
max_y = min_y + 1
# Create the mask
mask_roi = np.zeros((max_y - min_y, max_x - min_x), dtype=np.uint8)
cv2.fillPoly(mask_roi, [expanded_landmarks - [min_x, min_y]], 255)
# Apply Gaussian blur to soften the mask edges
mask_roi = cv2.GaussianBlur(mask_roi, (15, 15), 5)
# Place the mask ROI in the full-sized mask
mask[min_y:max_y, min_x:max_x] = mask_roi
# Extract the masked area from the frame
mouth_cutout = frame[min_y:max_y, min_x:max_x].copy()
# Return the expanded lower lip polygon in original frame coordinates
lower_lip_polygon = expanded_landmarks
return mask, mouth_cutout, (min_x, min_y, max_x, max_y), lower_lip_polygon
def draw_mouth_mask_visualization(
frame: Frame, face: Face, mouth_mask_data: tuple
) -> Frame:
landmarks = face.landmark_2d_106
if landmarks is not None and mouth_mask_data is not None:
mask, mouth_cutout, (min_x, min_y, max_x, max_y), lower_lip_polygon = (
mouth_mask_data
)
vis_frame = frame.copy()
# Ensure coordinates are within frame bounds
height, width = vis_frame.shape[:2]
min_x, min_y = max(0, min_x), max(0, min_y)
max_x, max_y = min(width, max_x), min(height, max_y)
# Adjust mask to match the region size
mask_region = mask[0 : max_y - min_y, 0 : max_x - min_x]
# Remove the color mask overlay
# color_mask = cv2.applyColorMap((mask_region * 255).astype(np.uint8), cv2.COLORMAP_JET)
# Ensure shapes match before blending
vis_region = vis_frame[min_y:max_y, min_x:max_x]
# Remove blending with color_mask
# if vis_region.shape[:2] == color_mask.shape[:2]:
# blended = cv2.addWeighted(vis_region, 0.7, color_mask, 0.3, 0)
# vis_frame[min_y:max_y, min_x:max_x] = blended
# Draw the lower lip polygon
cv2.polylines(vis_frame, [lower_lip_polygon], True, (0, 255, 0), 2)
# Remove the red box
# cv2.rectangle(vis_frame, (min_x, min_y), (max_x, max_y), (0, 0, 255), 2)
# Visualize the feathered mask
feather_amount = max(
1,
min(
30,
(max_x - min_x) // modules.globals.mask_feather_ratio,
(max_y - min_y) // modules.globals.mask_feather_ratio,
),
)
# Ensure kernel size is odd
kernel_size = 2 * feather_amount + 1
feathered_mask = cv2.GaussianBlur(
mask_region.astype(float), (kernel_size, kernel_size), 0
)
feathered_mask = (feathered_mask / feathered_mask.max() * 255).astype(np.uint8)
# Remove the feathered mask color overlay
# color_feathered_mask = cv2.applyColorMap(feathered_mask, cv2.COLORMAP_VIRIDIS)
# Ensure shapes match before blending feathered mask
# if vis_region.shape == color_feathered_mask.shape:
# blended_feathered = cv2.addWeighted(vis_region, 0.7, color_feathered_mask, 0.3, 0)
# vis_frame[min_y:max_y, min_x:max_x] = blended_feathered
# Add labels
cv2.putText(
vis_frame,
"Lower Mouth Mask",
(min_x, min_y - 10),
cv2.FONT_HERSHEY_SIMPLEX,
0.5,
(255, 255, 255),
1,
)
cv2.putText(
vis_frame,
"Feathered Mask",
(min_x, max_y + 20),
cv2.FONT_HERSHEY_SIMPLEX,
0.5,
(255, 255, 255),
1,
)
return vis_frame
return frame
def apply_mouth_area(
frame: np.ndarray,
mouth_cutout: np.ndarray,
mouth_box: tuple,
face_mask: np.ndarray,
mouth_polygon: np.ndarray,
) -> np.ndarray:
min_x, min_y, max_x, max_y = mouth_box
box_width = max_x - min_x
box_height = max_y - min_y
if (
mouth_cutout is None
or box_width is None
or box_height is None
or face_mask is None
or mouth_polygon is None
):
return frame
try:
resized_mouth_cutout = cv2.resize(mouth_cutout, (box_width, box_height))
roi = frame[min_y:max_y, min_x:max_x]
if roi.shape != resized_mouth_cutout.shape:
resized_mouth_cutout = cv2.resize(
resized_mouth_cutout, (roi.shape[1], roi.shape[0])
)
color_corrected_mouth = apply_color_transfer(resized_mouth_cutout, roi)
# Use the provided mouth polygon to create the mask
polygon_mask = np.zeros(roi.shape[:2], dtype=np.uint8)
adjusted_polygon = mouth_polygon - [min_x, min_y]
cv2.fillPoly(polygon_mask, [adjusted_polygon], 255)
# Apply feathering to the polygon mask
feather_amount = min(
30,
box_width // modules.globals.mask_feather_ratio,
box_height // modules.globals.mask_feather_ratio,
)
feathered_mask = cv2.GaussianBlur(
polygon_mask.astype(float), (0, 0), feather_amount
)
feathered_mask = feathered_mask / feathered_mask.max()
face_mask_roi = face_mask[min_y:max_y, min_x:max_x]
combined_mask = feathered_mask * (face_mask_roi / 255.0)
combined_mask = combined_mask[:, :, np.newaxis]
blended = (
color_corrected_mouth * combined_mask + roi * (1 - combined_mask)
).astype(np.uint8)
# Apply face mask to blended result
face_mask_3channel = (
np.repeat(face_mask_roi[:, :, np.newaxis], 3, axis=2) / 255.0
)
final_blend = blended * face_mask_3channel + roi * (1 - face_mask_3channel)
frame[min_y:max_y, min_x:max_x] = final_blend.astype(np.uint8)
except Exception as e:
pass
return frame
def create_face_mask(face: Face, frame: Frame) -> np.ndarray:
mask = np.zeros(frame.shape[:2], dtype=np.uint8)
landmarks = face.landmark_2d_106
if landmarks is not None:
# Convert landmarks to int32
landmarks = landmarks.astype(np.int32)
# Extract facial features
right_side_face = landmarks[0:16]
left_side_face = landmarks[17:32]
right_eye = landmarks[33:42]
right_eye_brow = landmarks[43:51]
left_eye = landmarks[87:96]
left_eye_brow = landmarks[97:105]
# Calculate forehead extension
right_eyebrow_top = np.min(right_eye_brow[:, 1])
left_eyebrow_top = np.min(left_eye_brow[:, 1])
eyebrow_top = min(right_eyebrow_top, left_eyebrow_top)
face_top = np.min([right_side_face[0, 1], left_side_face[-1, 1]])
forehead_height = face_top - eyebrow_top
extended_forehead_height = int(forehead_height * 5.0) # Extend by 50%
# Create forehead points
forehead_left = right_side_face[0].copy()
forehead_right = left_side_face[-1].copy()
forehead_left[1] -= extended_forehead_height
forehead_right[1] -= extended_forehead_height
# Combine all points to create the face outline
face_outline = np.vstack(
[
[forehead_left],
right_side_face,
left_side_face[
::-1
], # Reverse left side to create a continuous outline
[forehead_right],
]
)
# Calculate padding
padding = int(
np.linalg.norm(right_side_face[0] - left_side_face[-1]) * 0.05
) # 5% of face width
# Create a slightly larger convex hull for padding
hull = cv2.convexHull(face_outline)
hull_padded = []
for point in hull:
x, y = point[0]
center = np.mean(face_outline, axis=0)
direction = np.array([x, y]) - center
direction = direction / np.linalg.norm(direction)
padded_point = np.array([x, y]) + direction * padding
hull_padded.append(padded_point)
hull_padded = np.array(hull_padded, dtype=np.int32)
# Fill the padded convex hull
cv2.fillConvexPoly(mask, hull_padded, 255)
# Smooth the mask edges
mask = cv2.GaussianBlur(mask, (5, 5), 3)
return mask
def apply_color_transfer(source, target):
"""
Apply color transfer from target to source image
"""
source = cv2.cvtColor(source, cv2.COLOR_BGR2LAB).astype("float32")
target = cv2.cvtColor(target, cv2.COLOR_BGR2LAB).astype("float32")
source_mean, source_std = cv2.meanStdDev(source)
target_mean, target_std = cv2.meanStdDev(target)
# Reshape mean and std to be broadcastable
source_mean = source_mean.reshape(1, 1, 3)
source_std = source_std.reshape(1, 1, 3)
target_mean = target_mean.reshape(1, 1, 3)
target_std = target_std.reshape(1, 1, 3)
# Perform the color transfer
source = (source - source_mean) * (target_std / source_std) + target_mean
return cv2.cvtColor(np.clip(source, 0, 255).astype("uint8"), cv2.COLOR_LAB2BGR)
+145 -263
View File
@@ -3,7 +3,7 @@ import webbrowser
import customtkinter as ctk
from typing import Callable, Tuple
import cv2
from cv2_enumerate_cameras import enumerate_cameras
from cv2_enumerate_cameras import enumerate_cameras # Add this import
from PIL import Image, ImageOps
import time
import json
@@ -28,7 +28,6 @@ from modules.utilities import (
from modules.video_capture import VideoCapturer
from modules.gettext import LanguageManager
import platform
from modules.fake_face_handler import cleanup_fake_face, refresh_fake_face
if platform.system() == "Windows":
from pygrabber.dshow_graph import FilterGraph
@@ -36,7 +35,7 @@ if platform.system() == "Windows":
ROOT = None
POPUP = None
POPUP_LIVE = None
ROOT_HEIGHT = 730
ROOT_HEIGHT = 700
ROOT_WIDTH = 600
PREVIEW = None
@@ -79,12 +78,9 @@ target_label_dict_live = {}
img_ft, vid_ft = modules.globals.file_types
fake_face_switch = None
fake_face_value = None
def init(start: Callable[[], None], destroy: Callable[[], None], lang: str) -> ctk.CTk:
global ROOT, PREVIEW, _, fake_face_switch, fake_face_value
global ROOT, PREVIEW, _
lang_manager = LanguageManager(lang)
_ = lang_manager._
@@ -95,56 +91,51 @@ def init(start: Callable[[], None], destroy: Callable[[], None], lang: str) -> c
def save_switch_states():
try:
states = {
"keep_fps": modules.globals.keep_fps,
"keep_audio": modules.globals.keep_audio,
"keep_frames": modules.globals.keep_frames,
"many_faces": modules.globals.many_faces,
"map_faces": modules.globals.map_faces,
"color_correction": modules.globals.color_correction,
"nsfw_filter": modules.globals.nsfw_filter,
"live_mirror": modules.globals.live_mirror,
"live_resizable": modules.globals.live_resizable,
"fp_ui": modules.globals.fp_ui,
"show_fps": modules.globals.show_fps,
"mouth_mask": modules.globals.mouth_mask,
"show_mouth_mask_box": modules.globals.show_mouth_mask_box,
"use_fake_face": modules.globals.use_fake_face
}
with open(get_config_path(), 'w') as f:
json.dump(states, f)
except Exception as e:
print(f"Error saving switch states: {str(e)}")
switch_states = {
"keep_fps": modules.globals.keep_fps,
"keep_audio": modules.globals.keep_audio,
"keep_frames": modules.globals.keep_frames,
"many_faces": modules.globals.many_faces,
"map_faces": modules.globals.map_faces,
"color_correction": modules.globals.color_correction,
"nsfw_filter": modules.globals.nsfw_filter,
"live_mirror": modules.globals.live_mirror,
"live_resizable": modules.globals.live_resizable,
"fp_ui": modules.globals.fp_ui,
"show_fps": modules.globals.show_fps,
"mouth_mask": modules.globals.mouth_mask,
"show_mouth_mask_box": modules.globals.show_mouth_mask_box,
}
with open("switch_states.json", "w") as f:
json.dump(switch_states, f)
def load_switch_states():
try:
if os.path.exists(get_config_path()):
with open(get_config_path(), 'r') as f:
states = json.load(f)
modules.globals.keep_fps = states.get("keep_fps", True)
modules.globals.keep_audio = states.get("keep_audio", True)
modules.globals.keep_frames = states.get("keep_frames", False)
modules.globals.many_faces = states.get("many_faces", False)
modules.globals.map_faces = states.get("map_faces", False)
modules.globals.color_correction = states.get("color_correction", False)
modules.globals.nsfw_filter = states.get("nsfw_filter", False)
modules.globals.live_mirror = states.get("live_mirror", False)
modules.globals.live_resizable = states.get("live_resizable", False)
modules.globals.fp_ui = states.get("fp_ui", {"face_enhancer": False})
modules.globals.show_fps = states.get("show_fps", False)
modules.globals.mouth_mask = states.get("mouth_mask", False)
modules.globals.show_mouth_mask_box = states.get(
"show_mouth_mask_box", False
)
modules.globals.use_fake_face = False
except Exception as e:
print(f"Error loading switch states: {str(e)}")
with open("switch_states.json", "r") as f:
switch_states = json.load(f)
modules.globals.keep_fps = switch_states.get("keep_fps", True)
modules.globals.keep_audio = switch_states.get("keep_audio", True)
modules.globals.keep_frames = switch_states.get("keep_frames", False)
modules.globals.many_faces = switch_states.get("many_faces", False)
modules.globals.map_faces = switch_states.get("map_faces", False)
modules.globals.color_correction = switch_states.get("color_correction", False)
modules.globals.nsfw_filter = switch_states.get("nsfw_filter", False)
modules.globals.live_mirror = switch_states.get("live_mirror", False)
modules.globals.live_resizable = switch_states.get("live_resizable", False)
modules.globals.fp_ui = switch_states.get("fp_ui", {"face_enhancer": False})
modules.globals.show_fps = switch_states.get("show_fps", False)
modules.globals.mouth_mask = switch_states.get("mouth_mask", False)
modules.globals.show_mouth_mask_box = switch_states.get(
"show_mouth_mask_box", False
)
except FileNotFoundError:
# If the file doesn't exist, use default values
pass
def create_root(start: Callable[[], None], destroy: Callable[[], None]) -> ctk.CTk:
global source_label, target_label, status_label, show_fps_switch, fake_face_switch, fake_face_value
global source_label, target_label, status_label, show_fps_switch
load_switch_states()
@@ -159,28 +150,22 @@ def create_root(start: Callable[[], None], destroy: Callable[[], None]) -> ctk.C
)
root.configure()
root.protocol("WM_DELETE_WINDOW", lambda: destroy())
# Add icon to the main window
icon_path = resolve_relative_path("deeplivecam.ico")
if os.path.exists(icon_path):
root.iconbitmap(icon_path)
# Image Selection Area (Top)
source_label = ctk.CTkLabel(root, text=None)
source_label.place(relx=0.1, rely=0.05, relwidth=0.3, relheight=0.25)
source_label.place(relx=0.1, rely=0.1, relwidth=0.3, relheight=0.25)
target_label = ctk.CTkLabel(root, text=None)
target_label.place(relx=0.6, rely=0.05, relwidth=0.3, relheight=0.25)
target_label.place(relx=0.6, rely=0.1, relwidth=0.3, relheight=0.25)
select_face_button = ctk.CTkButton(
root, text=_("Select a face"), cursor="hand2", command=lambda: select_source_path()
)
select_face_button.place(relx=0.1, rely=0.35, relwidth=0.3, relheight=0.1)
select_face_button.place(relx=0.1, rely=0.4, relwidth=0.3, relheight=0.1)
swap_faces_button = ctk.CTkButton(
root, text="", cursor="hand2", command=lambda: swap_faces_paths()
)
swap_faces_button.place(relx=0.45, rely=0.35, relwidth=0.1, relheight=0.1)
swap_faces_button.place(relx=0.45, rely=0.4, relwidth=0.1, relheight=0.1)
select_target_button = ctk.CTkButton(
root,
@@ -188,30 +173,60 @@ def create_root(start: Callable[[], None], destroy: Callable[[], None]) -> ctk.C
cursor="hand2",
command=lambda: select_target_path(),
)
select_target_button.place(relx=0.6, rely=0.35, relwidth=0.3, relheight=0.1)
select_target_button.place(relx=0.6, rely=0.4, relwidth=0.3, relheight=0.1)
# AI Generated Face controls
fake_face_value = ctk.BooleanVar(value=modules.globals.use_fake_face)
fake_face_switch = ctk.CTkSwitch(
keep_fps_value = ctk.BooleanVar(value=modules.globals.keep_fps)
keep_fps_checkbox = ctk.CTkSwitch(
root,
text=_("Privacy Mode"),
variable=fake_face_value,
text=_("Keep fps"),
variable=keep_fps_value,
cursor="hand2",
command=lambda: toggle_fake_face(fake_face_value)
command=lambda: (
setattr(modules.globals, "keep_fps", keep_fps_value.get()),
save_switch_states(),
),
)
fake_face_switch.place(relx=0.1, rely=0.50)
keep_fps_checkbox.place(relx=0.1, rely=0.6)
# Add refresh button next to the switch
refresh_face_button = ctk.CTkButton(
keep_frames_value = ctk.BooleanVar(value=modules.globals.keep_frames)
keep_frames_switch = ctk.CTkSwitch(
root,
text="",
width=30,
text=_("Keep frames"),
variable=keep_frames_value,
cursor="hand2",
command=lambda: refresh_fake_face_clicked()
command=lambda: (
setattr(modules.globals, "keep_frames", keep_frames_value.get()),
save_switch_states(),
),
)
refresh_face_button.place(relx=0.35, rely=0.50)
keep_frames_switch.place(relx=0.1, rely=0.65)
enhancer_value = ctk.BooleanVar(value=modules.globals.fp_ui["face_enhancer"])
enhancer_switch = ctk.CTkSwitch(
root,
text=_("Face Enhancer"),
variable=enhancer_value,
cursor="hand2",
command=lambda: (
update_tumbler("face_enhancer", enhancer_value.get()),
save_switch_states(),
),
)
enhancer_switch.place(relx=0.1, rely=0.7)
keep_audio_value = ctk.BooleanVar(value=modules.globals.keep_audio)
keep_audio_switch = ctk.CTkSwitch(
root,
text=_("Keep audio"),
variable=keep_audio_value,
cursor="hand2",
command=lambda: (
setattr(modules.globals, "keep_audio", keep_audio_value.get()),
save_switch_states(),
),
)
keep_audio_switch.place(relx=0.6, rely=0.6)
# Face Processing Options (Middle Left)
many_faces_value = ctk.BooleanVar(value=modules.globals.many_faces)
many_faces_switch = ctk.CTkSwitch(
root,
@@ -223,7 +238,24 @@ def create_root(start: Callable[[], None], destroy: Callable[[], None]) -> ctk.C
save_switch_states(),
),
)
many_faces_switch.place(relx=0.1, rely=0.55)
many_faces_switch.place(relx=0.6, rely=0.65)
color_correction_value = ctk.BooleanVar(value=modules.globals.color_correction)
color_correction_switch = ctk.CTkSwitch(
root,
text=_("Fix Blueish Cam"),
variable=color_correction_value,
cursor="hand2",
command=lambda: (
setattr(modules.globals, "color_correction", color_correction_value.get()),
save_switch_states(),
),
)
color_correction_switch.place(relx=0.6, rely=0.70)
# nsfw_value = ctk.BooleanVar(value=modules.globals.nsfw_filter)
# nsfw_switch = ctk.CTkSwitch(root, text='NSFW filter', variable=nsfw_value, cursor='hand2', command=lambda: setattr(modules.globals, 'nsfw_filter', nsfw_value.get()))
# nsfw_switch.place(relx=0.6, rely=0.7)
map_faces = ctk.BooleanVar(value=modules.globals.map_faces)
map_faces_switch = ctk.CTkSwitch(
@@ -237,35 +269,8 @@ def create_root(start: Callable[[], None], destroy: Callable[[], None]) -> ctk.C
close_mapper_window() if not map_faces.get() else None
),
)
map_faces_switch.place(relx=0.1, rely=0.60)
map_faces_switch.place(relx=0.1, rely=0.75)
enhancer_value = ctk.BooleanVar(value=modules.globals.fp_ui["face_enhancer"])
enhancer_switch = ctk.CTkSwitch(
root,
text=_("Face Enhancer"),
variable=enhancer_value,
cursor="hand2",
command=lambda: (
update_tumbler("face_enhancer", enhancer_value.get()),
save_switch_states(),
),
)
enhancer_switch.place(relx=0.1, rely=0.65)
keep_audio_value = ctk.BooleanVar(value=modules.globals.keep_audio)
keep_audio_switch = ctk.CTkSwitch(
root,
text=_("Keep audio"),
variable=keep_audio_value,
cursor="hand2",
command=lambda: (
setattr(modules.globals, "keep_audio", keep_audio_value.get()),
save_switch_states(),
),
)
keep_audio_switch.place(relx=0.1, rely=0.70)
# Add show FPS switch right after keep_audio_switch
show_fps_value = ctk.BooleanVar(value=modules.globals.show_fps)
show_fps_switch = ctk.CTkSwitch(
root,
@@ -277,9 +282,8 @@ def create_root(start: Callable[[], None], destroy: Callable[[], None]) -> ctk.C
save_switch_states(),
),
)
show_fps_switch.place(relx=0.1, rely=0.75)
show_fps_switch.place(relx=0.6, rely=0.75)
# Mask Switches (Middle Right - Top Section)
mouth_mask_var = ctk.BooleanVar(value=modules.globals.mouth_mask)
mouth_mask_switch = ctk.CTkSwitch(
root,
@@ -288,117 +292,38 @@ def create_root(start: Callable[[], None], destroy: Callable[[], None]) -> ctk.C
cursor="hand2",
command=lambda: setattr(modules.globals, "mouth_mask", mouth_mask_var.get()),
)
mouth_mask_switch.place(relx=0.6, rely=0.50)
mouth_mask_switch.place(relx=0.1, rely=0.55)
# Add mouth mask size slider
mouth_mask_size_slider = ctk.CTkSlider(
root,
from_=0.5,
to=2.0,
number_of_steps=30,
command=lambda value: setattr(modules.globals, "mouth_mask_size", value)
)
mouth_mask_size_slider.set(modules.globals.mouth_mask_size)
mouth_mask_size_slider.place(relx=0.8, rely=0.50, relwidth=0.1)
eyes_mask_var = ctk.BooleanVar(value=modules.globals.eyes_mask)
eyes_mask_switch = ctk.CTkSwitch(
root,
text=_("Eyes Mask"),
variable=eyes_mask_var,
cursor="hand2",
command=lambda: setattr(modules.globals, "eyes_mask", eyes_mask_var.get()),
)
eyes_mask_switch.place(relx=0.6, rely=0.55)
# Add eyes mask size slider
eyes_mask_size_slider = ctk.CTkSlider(
root,
from_=0.5,
to=2.0,
number_of_steps=30,
command=lambda value: setattr(modules.globals, "eyes_mask_size", value)
)
eyes_mask_size_slider.set(modules.globals.eyes_mask_size)
eyes_mask_size_slider.place(relx=0.8, rely=0.55, relwidth=0.1)
eyebrows_mask_var = ctk.BooleanVar(value=modules.globals.eyebrows_mask)
eyebrows_mask_switch = ctk.CTkSwitch(
root,
text=_("Eyebrows Mask"),
variable=eyebrows_mask_var,
cursor="hand2",
command=lambda: setattr(modules.globals, "eyebrows_mask", eyebrows_mask_var.get()),
)
eyebrows_mask_switch.place(relx=0.6, rely=0.60)
# Add eyebrows mask size slider
eyebrows_mask_size_slider = ctk.CTkSlider(
root,
from_=0.5,
to=2.0,
number_of_steps=30,
command=lambda value: setattr(modules.globals, "eyebrows_mask_size", value)
)
eyebrows_mask_size_slider.set(modules.globals.eyebrows_mask_size)
eyebrows_mask_size_slider.place(relx=0.8, rely=0.60, relwidth=0.1)
# Box Visualization Switches (Middle Right - Bottom Section)
show_mouth_mask_box_var = ctk.BooleanVar(value=modules.globals.show_mouth_mask_box)
show_mouth_mask_box_switch = ctk.CTkSwitch(
root,
text=_("Show Mouth Box"),
text=_("Show Mouth Mask Box"),
variable=show_mouth_mask_box_var,
cursor="hand2",
command=lambda: setattr(
modules.globals, "show_mouth_mask_box", show_mouth_mask_box_var.get()
),
)
show_mouth_mask_box_switch.place(relx=0.6, rely=0.65)
show_mouth_mask_box_switch.place(relx=0.6, rely=0.55)
show_eyes_mask_box_var = ctk.BooleanVar(value=modules.globals.show_eyes_mask_box)
show_eyes_mask_box_switch = ctk.CTkSwitch(
root,
text=_("Show Eyes Box"),
variable=show_eyes_mask_box_var,
cursor="hand2",
command=lambda: setattr(
modules.globals, "show_eyes_mask_box", show_eyes_mask_box_var.get()
),
)
show_eyes_mask_box_switch.place(relx=0.6, rely=0.70)
show_eyebrows_mask_box_var = ctk.BooleanVar(value=modules.globals.show_eyebrows_mask_box)
show_eyebrows_mask_box_switch = ctk.CTkSwitch(
root,
text=_("Show Eyebrows Box"),
variable=show_eyebrows_mask_box_var,
cursor="hand2",
command=lambda: setattr(
modules.globals, "show_eyebrows_mask_box", show_eyebrows_mask_box_var.get()
),
)
show_eyebrows_mask_box_switch.place(relx=0.6, rely=0.75)
# Main Control Buttons (Bottom)
start_button = ctk.CTkButton(
root, text=_("Start"), cursor="hand2", command=lambda: analyze_target(start, root)
)
start_button.place(relx=0.15, rely=0.80, relwidth=0.2, relheight=0.05)
preview_button = ctk.CTkButton(
root, text=_("Preview"), cursor="hand2", command=lambda: toggle_preview()
)
preview_button.place(relx=0.4, rely=0.80, relwidth=0.2, relheight=0.05)
stop_button = ctk.CTkButton(
root, text=_("Destroy"), cursor="hand2", command=lambda: destroy()
)
stop_button.place(relx=0.65, rely=0.80, relwidth=0.2, relheight=0.05)
stop_button.place(relx=0.4, rely=0.80, relwidth=0.2, relheight=0.05)
# Camera Section (Bottom)
preview_button = ctk.CTkButton(
root, text=_("Preview"), cursor="hand2", command=lambda: toggle_preview()
)
preview_button.place(relx=0.65, rely=0.80, relwidth=0.2, relheight=0.05)
# --- Camera Selection ---
camera_label = ctk.CTkLabel(root, text=_("Select Camera:"))
camera_label.place(relx=0.1, rely=0.87, relwidth=0.2, relheight=0.05)
camera_label.place(relx=0.1, rely=0.86, relwidth=0.2, relheight=0.05)
available_cameras = get_available_cameras()
camera_indices, camera_names = available_cameras
@@ -417,7 +342,7 @@ def create_root(start: Callable[[], None], destroy: Callable[[], None]) -> ctk.C
root, variable=camera_variable, values=camera_names
)
camera_optionmenu.place(relx=0.35, rely=0.87, relwidth=0.25, relheight=0.05)
camera_optionmenu.place(relx=0.35, rely=0.86, relwidth=0.25, relheight=0.05)
live_button = ctk.CTkButton(
root,
@@ -437,16 +362,16 @@ def create_root(start: Callable[[], None], destroy: Callable[[], None]) -> ctk.C
else "disabled"
),
)
live_button.place(relx=0.65, rely=0.87, relwidth=0.2, relheight=0.05)
live_button.place(relx=0.65, rely=0.86, relwidth=0.2, relheight=0.05)
# --- End Camera Selection ---
# Status and Links (Bottom)
status_label = ctk.CTkLabel(root, text=None, justify="center")
status_label.place(relx=0.1, rely=0.92, relwidth=0.8)
status_label.place(relx=0.1, rely=0.9, relwidth=0.8)
donate_label = ctk.CTkLabel(
root, text="Deep Live Cam", justify="center", cursor="hand2"
)
donate_label.place(relx=0.1, rely=0.94, relwidth=0.8)
donate_label.place(relx=0.1, rely=0.95, relwidth=0.8)
donate_label.configure(
text_color=ctk.ThemeManager.theme.get("URL").get("text_color")
)
@@ -472,7 +397,7 @@ def analyze_target(start: Callable[[], None], root: ctk.CTk):
return
if modules.globals.map_faces:
modules.globals.souce_target_map = []
modules.globals.source_target_map = []
if is_image(modules.globals.target_path):
update_status("Getting unique faces")
@@ -481,8 +406,8 @@ def analyze_target(start: Callable[[], None], root: ctk.CTk):
update_status("Getting unique faces")
get_unique_faces_from_target_video()
if len(modules.globals.souce_target_map) > 0:
create_source_target_popup(start, root, modules.globals.souce_target_map)
if len(modules.globals.source_target_map) > 0:
create_source_target_popup(start, root, modules.globals.source_target_map)
else:
update_status("No faces found in target")
else:
@@ -619,11 +544,6 @@ def create_preview(parent: ctk.CTkToplevel) -> ctk.CTkToplevel:
preview.configure()
preview.protocol("WM_DELETE_WINDOW", lambda: toggle_preview())
preview.resizable(width=True, height=True)
# Add icon to the preview window
icon_path = resolve_relative_path("deeplivecam.ico")
if os.path.exists(icon_path):
preview.iconbitmap(icon_path)
preview_label = ctk.CTkLabel(preview, text=None)
preview_label.pack(fill="both", expand=True)
@@ -660,7 +580,7 @@ def update_tumbler(var: str, value: bool) -> None:
def select_source_path() -> None:
global RECENT_DIRECTORY_SOURCE, img_ft, vid_ft, fake_face_switch, fake_face_value
global RECENT_DIRECTORY_SOURCE, img_ft, vid_ft
PREVIEW.withdraw()
source_path = ctk.filedialog.askopenfilename(
@@ -669,10 +589,6 @@ def select_source_path() -> None:
filetypes=[img_ft],
)
if is_image(source_path):
modules.globals.use_fake_face = False
fake_face_value.set(False)
cleanup_fake_face()
modules.globals.source_path = source_path
RECENT_DIRECTORY_SOURCE = os.path.dirname(modules.globals.source_path)
image = render_image_preview(modules.globals.source_path, (200, 200))
@@ -780,17 +696,21 @@ def check_and_ignore_nsfw(target, destroy: Callable = None) -> bool:
def fit_image_to_size(image, width: int, height: int):
if width is None and height is None:
if width is None or height is None or width <= 0 or height <= 0:
return image
h, w, _ = image.shape
ratio_h = 0.0
ratio_w = 0.0
if width > height:
ratio_h = height / h
else:
ratio_w = width / w
ratio = max(ratio_w, ratio_h)
new_size = (int(ratio * w), int(ratio * h))
ratio_w = width / w
ratio_h = height / h
# Use the smaller ratio to ensure the image fits within the given dimensions
ratio = min(ratio_w, ratio_h)
# Compute new dimensions, ensuring they're at least 1 pixel
new_width = max(1, int(ratio * w))
new_height = max(1, int(ratio * h))
new_size = (new_width, new_height)
return cv2.resize(image, dsize=new_size)
@@ -845,7 +765,8 @@ def update_preview(frame_number: int = 0) -> None:
modules.globals.frame_processors
):
temp_frame = frame_processor.process_frame(
get_one_face(cv2.imread(modules.globals.source_path)), temp_frame)
get_one_face(cv2.imread(modules.globals.source_path)), temp_frame
)
image = Image.fromarray(cv2.cvtColor(temp_frame, cv2.COLOR_BGR2RGB))
image = ImageOps.contain(
image, (PREVIEW_MAX_WIDTH, PREVIEW_MAX_HEIGHT), Image.LANCZOS
@@ -870,9 +791,9 @@ def webcam_preview(root: ctk.CTk, camera_index: int):
return
create_webcam_preview(camera_index)
else:
modules.globals.souce_target_map = []
modules.globals.source_target_map = []
create_source_target_popup_for_webcam(
root, modules.globals.souce_target_map, camera_index
root, modules.globals.source_target_map, camera_index
)
@@ -1283,42 +1204,3 @@ def update_webcam_target(
else:
update_pop_live_status("Face could not be detected in last upload!")
return map
def toggle_fake_face(switch_var: ctk.BooleanVar) -> None:
modules.globals.use_fake_face = switch_var.get()
if modules.globals.use_fake_face:
if not modules.globals.fake_face_path:
if refresh_fake_face():
modules.globals.source_path = modules.globals.fake_face_path
# Update the source image preview
image = render_image_preview(modules.globals.source_path, (200, 200))
source_label.configure(image=image)
else:
cleanup_fake_face()
# Clear the source image preview
source_label.configure(image=None)
modules.globals.source_path = None
def refresh_fake_face_clicked() -> None:
"""Handle refresh button click to update fake face during live preview"""
if not modules.globals.use_fake_face:
# If privacy mode is off, turn it on first
modules.globals.use_fake_face = True
fake_face_value.set(True)
if refresh_fake_face():
modules.globals.source_path = modules.globals.fake_face_path
# Update the source image preview
image = render_image_preview(modules.globals.source_path, (200, 200))
source_label.configure(image=image)
# Force reload of frame processors to use new source face
global FRAME_PROCESSORS_MODULES
FRAME_PROCESSORS_MODULES = []
frame_processors = get_frame_processors_modules(modules.globals.frame_processors)
def get_config_path() -> str:
"""Get the path to the config file"""
config_dir = os.path.join(os.path.expanduser("~"), ".deep-live-cam")
os.makedirs(config_dir, exist_ok=True)
return os.path.join(config_dir, "switch_states.json")
+6 -5
View File
@@ -1,6 +1,7 @@
--extra-index-url https://download.pytorch.org/whl/cu118
numpy>=1.23.5,<2
typing-extensions>=4.8.0
opencv-python==4.10.0.84
cv2_enumerate_cameras==1.1.15
onnx==1.16.0
@@ -9,13 +10,13 @@ psutil==5.9.8
tk==0.1.0
customtkinter==5.2.2
pillow==11.1.0
torch==2.0.1+cu118; sys_platform != 'darwin'
torch==2.0.1; sys_platform == 'darwin'
torchvision==0.15.2+cu118; sys_platform != 'darwin'
torchvision==0.15.2; sys_platform == 'darwin'
torch==2.5.1+cu118; sys_platform != 'darwin'
torch==2.5.1; sys_platform == 'darwin'
torchvision==0.20.1; sys_platform != 'darwin'
torchvision==0.20.1; sys_platform == 'darwin'
onnxruntime-silicon==1.16.3; sys_platform == 'darwin' and platform_machine == 'arm64'
onnxruntime-gpu==1.16.3; sys_platform != 'darwin'
tensorflow==2.12.1; sys_platform != 'darwin'
tensorflow; sys_platform != 'darwin'
opennsfw2==0.10.2
protobuf==4.23.2
tqdm==4.66.4