diff --git a/cli/api-txt2img.py b/cli/api-txt2img.py
index fe085e7f1..89d84be80 100755
--- a/cli/api-txt2img.py
+++ b/cli/api-txt2img.py
@@ -49,7 +49,7 @@ def generate(args): # pylint: disable=redefined-outer-name
options['width'] = int(args.width)
options['height'] = int(args.height)
if args.faces:
- options['restore_faces'] = args.faces
+ options['detailer'] = args.detailer
options['denoising_strength'] = 0.5
options['hr_sampler_name'] = args.sampler
data = post('/sdapi/v1/txt2img', options)
@@ -75,7 +75,7 @@ if __name__ == "__main__":
parser.add_argument('--height', required=False, default=512, help='image height')
parser.add_argument('--steps', required=False, default=20, help='number of steps')
parser.add_argument('--seed', required=False, default=-1, help='initial seed')
- parser.add_argument('--faces', action='store_true', help='restore faces')
+ parser.add_argument('--detailer', action='store_true', help='run detailer')
parser.add_argument('--sampler', required=False, default='Euler a', help='sampler name')
parser.add_argument('--output', required=False, default=None, help='output image file')
parser.add_argument('--model', required=False, help='model name')
diff --git a/cli/create-previews.py b/cli/create-previews.py
index 58c2136df..fd1d81a7f 100755
--- a/cli/create-previews.py
+++ b/cli/create-previews.py
@@ -46,7 +46,7 @@ options = Map({
},
# generate params
'generate': {
- 'restore_faces': True,
+ 'detailer': True,
'prompt': '',
'negative_prompt': 'foggy, blurry, blurred, duplicate, ugly, mutilated, mutation, mutated, out of frame, bad anatomy, disfigured, deformed, censored, low res, low resolution, watermark, text, poorly drawn face, poorly drawn hands, signature',
'steps': 20,
diff --git a/cli/generate.json b/cli/generate.json
index 0bbcb1439..c379a6f64 100644
--- a/cli/generate.json
+++ b/cli/generate.json
@@ -8,7 +8,7 @@
},
"generate":
{
- "restore_faces": true,
+ "detailer": true,
"prompt": "dynamic",
"negative_prompt": "foggy, blurry, blurred, duplicate, ugly, mutilated, mutation, mutated, out of frame, bad anatomy, disfigured, deformed, censored, low res, watermark, text, poorly drawn face, signature",
"steps": 30,
diff --git a/cli/generate.py b/cli/generate.py
index b8d5cac66..9b784373a 100755
--- a/cli/generate.py
+++ b/cli/generate.py
@@ -230,7 +230,7 @@ def args(): # parse cmd arguments
parser.add_argument('--style', type = str, default = 'random', required = False, help = 'image style, used to guide dynamic prompt when prompt is not provided')
parser.add_argument('--suffix', type = str, default = 'random', required = False, help = 'style suffix, used to guide dynamic prompt when prompt is not provided')
parser.add_argument('--place', type = str, default = 'random', required = False, help = 'place locator, used to guide dynamic prompt when prompt is not provided')
- parser.add_argument('--faces', default = False, action='store_true', help = 'restore faces during upscaling')
+ parser.add_argument('--detailer', default = False, action='store_true', help = 'run detailer')
parser.add_argument('--steps', type = int, default = 0, required = False, help = 'number of steps')
parser.add_argument('--batch', type = int, default = 0, required = False, help = 'batch size, limited by gpu vram')
parser.add_argument('--n', type = int, default = 0, required = False, help = 'number of iterations')
@@ -299,7 +299,7 @@ def args(): # parse cmd arguments
_dynamic = prompt(params)
sd.paths.root = params.path if params.path != '' else sd.paths.root
- sd.generate.restore_faces = params.faces if params.faces is not None else sd.generate.restore_faces
+ sd.generate.detailer = params.detailer if params.detailer is not None else sd.generate.detailer
sd.generate.seed = params.seed if params.seed > 0 else sd.generate.seed
sd.generate.sampler_name = params.sampler if params.sampler != 'random' else sd.generate.sampler_name
sd.generate.batch_size = params.batch if params.batch > 0 else sd.generate.batch_size
@@ -309,7 +309,7 @@ def args(): # parse cmd arguments
sd.generate.height = params.height if params.height > 0 else sd.generate.height
sd.generate.steps = params.steps if params.steps > 0 else sd.generate.steps
sd.upscale.upscaling_resize = params.upscale if params.upscale > 0 else sd.upscale.upscaling_resize
- sd.upscale.codeformer_visibility = 1 if params.faces else sd.upscale.codeformer_visibility
+ sd.upscale.codeformer_visibility = 1 if params.detailer else sd.upscale.codeformer_visibility
sd.options.sd_vae = params.vae if params.vae != '' else sd.options.sd_vae
sd.options.sd_model_checkpoint = params.model if params.model != '' else sd.options.sd_model_checkpoint
sd.upscale.upscaler_1 = 'SwinIR_4x' if params.upscale > 1 else sd.upscale.upscaler_1
diff --git a/extensions-builtin/sd-extension-system-info b/extensions-builtin/sd-extension-system-info
index c88e83d40..5267e3264 160000
--- a/extensions-builtin/sd-extension-system-info
+++ b/extensions-builtin/sd-extension-system-info
@@ -1 +1 @@
-Subproject commit c88e83d403e1cae478df870fa2dd277d2028dc34
+Subproject commit 5267e326499cb996c8804a02e8844a2f6d77a0d4
diff --git a/javascript/contextMenus.js b/javascript/contextMenus.js
index a0424f266..b15615a2a 100644
--- a/javascript/contextMenus.js
+++ b/javascript/contextMenus.js
@@ -129,7 +129,7 @@ async function initContextMenu() {
id = `#${tab}_reprocess`;
appendContextMenuOption(id, 'Decode full quality', () => reprocessClick(`${tab}`, 'reprocess_decode'), true);
appendContextMenuOption(id, 'Refine & HiRes pass', () => reprocessClick(`${tab}`, 'reprocess_refine'), true);
- appendContextMenuOption(id, 'Face restore', () => reprocessClick(`${tab}`, 'reprocess_face'), true);
+ appendContextMenuOption(id, 'Detailer pass', () => reprocessClick(`${tab}`, 'reprocess_detail'), true);
}
addContextMenuEventListener();
}
diff --git a/modules/api/api.py b/modules/api/api.py
index 890a18733..4b7c750fd 100644
--- a/modules/api/api.py
+++ b/modules/api/api.py
@@ -69,7 +69,7 @@ class Api:
self.add_api_route("/sdapi/v1/upscalers", endpoints.get_upscalers, methods=["GET"], response_model=List[models.ItemUpscaler])
self.add_api_route("/sdapi/v1/sd-models", endpoints.get_sd_models, methods=["GET"], response_model=List[models.ItemModel])
self.add_api_route("/sdapi/v1/hypernetworks", endpoints.get_hypernetworks, methods=["GET"], response_model=List[models.ItemHypernetwork])
- self.add_api_route("/sdapi/v1/face-restorers", endpoints.get_face_restorers, methods=["GET"], response_model=List[models.ItemFaceRestorer])
+ self.add_api_route("/sdapi/v1/face-restorers", endpoints.get_detailers, methods=["GET"], response_model=List[models.ItemDetailer])
self.add_api_route("/sdapi/v1/prompt-styles", endpoints.get_prompt_styles, methods=["GET"], response_model=List[models.ItemStyle])
self.add_api_route("/sdapi/v1/embeddings", endpoints.get_embeddings, methods=["GET"], response_model=models.ResEmbeddings)
self.add_api_route("/sdapi/v1/sd-vae", endpoints.get_sd_vaes, methods=["GET"], response_model=List[models.ItemVae])
diff --git a/modules/api/endpoints.py b/modules/api/endpoints.py
index 63c5764c0..310abed1e 100644
--- a/modules/api/endpoints.py
+++ b/modules/api/endpoints.py
@@ -23,8 +23,8 @@ def get_sd_models():
def get_hypernetworks():
return [{"name": name, "path": shared.hypernetworks[name]} for name in shared.hypernetworks]
-def get_face_restorers():
- return [{"name":x.name(), "cmd_dir": getattr(x, "cmd_dir", None)} for x in shared.face_restorers]
+def get_detailers():
+ return [{"name":x.name(), "cmd_dir": getattr(x, "cmd_dir", None)} for x in shared.detailers]
def get_prompt_styles():
return [{ 'name': v.name, 'prompt': v.prompt, 'negative_prompt': v.negative_prompt, 'extra': v.extra, 'filename': v.filename, 'preview': v.preview} for v in shared.prompt_styles.styles.values()]
diff --git a/modules/api/models.py b/modules/api/models.py
index 3c5fd6146..6e658b484 100644
--- a/modules/api/models.py
+++ b/modules/api/models.py
@@ -110,7 +110,7 @@ class ItemHypernetwork(BaseModel):
name: str = Field(title="Name")
path: Optional[str] = Field(title="Path")
-class ItemFaceRestorer(BaseModel):
+class ItemDetailer(BaseModel):
name: str = Field(title="Name")
cmd_dir: Optional[str] = Field(title="Path")
diff --git a/modules/api/process.py b/modules/api/process.py
index 6dd5e701e..505ba1374 100644
--- a/modules/api/process.py
+++ b/modules/api/process.py
@@ -107,7 +107,7 @@ class APIProcess():
return ResMask(mask=image)
def post_face(self, req: ReqFace):
- from scripts.face_details import yolo # pylint: disable=no-name-in-module
+ from scripts.detailer import yolo # pylint: disable=no-name-in-module
image = decode_base64_to_image(req.image)
shared.state.begin('API-FACE', api=True)
images = []
diff --git a/modules/control/run.py b/modules/control/run.py
index 4f8b48cd2..01779119e 100644
--- a/modules/control/run.py
+++ b/modules/control/run.py
@@ -60,7 +60,7 @@ def control_run(state: str = '',
steps: int = 20, sampler_index: int = None,
seed: int = -1, subseed: int = -1, subseed_strength: float = 0, seed_resize_from_h: int = -1, seed_resize_from_w: int = -1,
cfg_scale: float = 6.0, clip_skip: float = 1.0, image_cfg_scale: float = 6.0, diffusers_guidance_rescale: float = 0.7, pag_scale: float = 0.0, pag_adaptive: float = 0.5, cfg_end: float = 1.0,
- full_quality: bool = True, restore_faces: bool = False, tiling: bool = False, hidiffusion: bool = False,
+ full_quality: bool = True, detailer: bool = False, tiling: bool = False, hidiffusion: bool = False,
hdr_mode: int = 0, hdr_brightness: float = 0, hdr_color: float = 0, hdr_sharpen: float = 0, hdr_clamp: bool = False, hdr_boundary: float = 4.0, hdr_threshold: float = 0.95,
hdr_maximize: bool = False, hdr_max_center: float = 0.6, hdr_max_boundry: float = 1.0, hdr_color_picker: str = None, hdr_tint_ratio: float = 0,
resize_mode_before: int = 0, resize_name_before: str = 'None', resize_context_before: str = 'None', width_before: int = 512, height_before: int = 512, scale_by_before: float = 1.0, selected_scale_tab_before: int = 0,
@@ -129,7 +129,7 @@ def control_run(state: str = '',
pag_scale = pag_scale,
pag_adaptive = pag_adaptive,
full_quality = full_quality,
- restore_faces = restore_faces,
+ detailer = detailer,
tiling = tiling,
hidiffusion = hidiffusion,
# resize
diff --git a/modules/detailer.py b/modules/detailer.py
new file mode 100644
index 000000000..86eae2230
--- /dev/null
+++ b/modules/detailer.py
@@ -0,0 +1,17 @@
+from modules import shared
+
+
+class Detailer:
+ def name(self):
+ return "None"
+
+ def restore(self, np_image):
+ return np_image
+
+
+def detail(np_image, p=None):
+ detailers = [x for x in shared.detailers if x.name() == shared.opts.detailer_model or shared.opts.detailer_model is None]
+ if len(detailers) == 0:
+ return np_image
+ detailer = detailers[0]
+ return detailer.restore(np_image, p)
diff --git a/modules/face/__init__.py b/modules/face/__init__.py
index d5d51e647..aae3cdaa2 100644
--- a/modules/face/__init__.py
+++ b/modules/face/__init__.py
@@ -134,7 +134,7 @@ class Script(scripts.Script):
from modules.face.insightface import get_app
app=get_app('buffalo_l')
from modules.face.faceswap import face_swap
- if shared.opts.save_images_before_face_restoration and not p.do_not_save_samples:
+ if shared.opts.save_images_before_detailer and not p.do_not_save_samples:
for i, image in enumerate(processed.images):
info = processing.create_infotext(p, index=i)
images.save_image(image, path=p.outpath_samples, seed=p.all_seeds[i], prompt=p.all_prompts[i], info=info, p=p, suffix="-before-faceswap")
diff --git a/modules/face_restoration.py b/modules/face_restoration.py
deleted file mode 100644
index d17191fdf..000000000
--- a/modules/face_restoration.py
+++ /dev/null
@@ -1,17 +0,0 @@
-from modules import shared
-
-
-class FaceRestoration:
- def name(self):
- return "None"
-
- def restore(self, np_image):
- return np_image
-
-
-def restore_faces(np_image, p=None):
- face_restorers = [x for x in shared.face_restorers if x.name() == shared.opts.face_restoration_model or shared.opts.face_restoration_model is None]
- if len(face_restorers) == 0:
- return np_image
- face_restorer = face_restorers[0]
- return face_restorer.restore(np_image, p)
diff --git a/modules/img2img.py b/modules/img2img.py
index 6e2986608..2417c01b3 100644
--- a/modules/img2img.py
+++ b/modules/img2img.py
@@ -120,7 +120,7 @@ def img2img(id_task: str, state: str, mode: int,
sampler_index,
mask_blur, mask_alpha,
inpainting_fill,
- full_quality, restore_faces, tiling, hidiffusion,
+ full_quality, detailer, tiling, hidiffusion,
n_iter, batch_size,
cfg_scale, image_cfg_scale,
diffusers_guidance_rescale,
@@ -144,7 +144,7 @@ def img2img(id_task: str, state: str, mode: int,
shared.log.warning('Model not loaded')
return [], '', '', 'Error: model not loaded'
- debug(f'img2img: id_task={id_task}|mode={mode}|prompt={prompt}|negative_prompt={negative_prompt}|prompt_styles={prompt_styles}|init_img={init_img}|sketch={sketch}|init_img_with_mask={init_img_with_mask}|inpaint_color_sketch={inpaint_color_sketch}|inpaint_color_sketch_orig={inpaint_color_sketch_orig}|init_img_inpaint={init_img_inpaint}|init_mask_inpaint={init_mask_inpaint}|steps={steps}|sampler_index={sampler_index}||mask_blur={mask_blur}|mask_alpha={mask_alpha}|inpainting_fill={inpainting_fill}|full_quality={full_quality}|restore_faces={restore_faces}|tiling={tiling}|hidiffusion={hidiffusion}|n_iter={n_iter}|batch_size={batch_size}|cfg_scale={cfg_scale}|image_cfg_scale={image_cfg_scale}|clip_skip={clip_skip}|denoising_strength={denoising_strength}|seed={seed}|subseed{subseed}|subseed_strength={subseed_strength}|seed_resize_from_h={seed_resize_from_h}|seed_resize_from_w={seed_resize_from_w}|selected_scale_tab={selected_scale_tab}|height={height}|width={width}|scale_by={scale_by}|resize_mode={resize_mode}|resize_name={resize_name}|resize_context={resize_context}|inpaint_full_res={inpaint_full_res}|inpaint_full_res_padding={inpaint_full_res_padding}|inpainting_mask_invert={inpainting_mask_invert}|img2img_batch_files={img2img_batch_files}|img2img_batch_input_dir={img2img_batch_input_dir}|img2img_batch_output_dir={img2img_batch_output_dir}|img2img_batch_inpaint_mask_dir={img2img_batch_inpaint_mask_dir}|override_settings_texts={override_settings_texts}')
+ debug(f'img2img: id_task={id_task}|mode={mode}|prompt={prompt}|negative_prompt={negative_prompt}|prompt_styles={prompt_styles}|init_img={init_img}|sketch={sketch}|init_img_with_mask={init_img_with_mask}|inpaint_color_sketch={inpaint_color_sketch}|inpaint_color_sketch_orig={inpaint_color_sketch_orig}|init_img_inpaint={init_img_inpaint}|init_mask_inpaint={init_mask_inpaint}|steps={steps}|sampler_index={sampler_index}||mask_blur={mask_blur}|mask_alpha={mask_alpha}|inpainting_fill={inpainting_fill}|full_quality={full_quality}|detailer={detailer}|tiling={tiling}|hidiffusion={hidiffusion}|n_iter={n_iter}|batch_size={batch_size}|cfg_scale={cfg_scale}|image_cfg_scale={image_cfg_scale}|clip_skip={clip_skip}|denoising_strength={denoising_strength}|seed={seed}|subseed{subseed}|subseed_strength={subseed_strength}|seed_resize_from_h={seed_resize_from_h}|seed_resize_from_w={seed_resize_from_w}|selected_scale_tab={selected_scale_tab}|height={height}|width={width}|scale_by={scale_by}|resize_mode={resize_mode}|resize_name={resize_name}|resize_context={resize_context}|inpaint_full_res={inpaint_full_res}|inpaint_full_res_padding={inpaint_full_res_padding}|inpainting_mask_invert={inpainting_mask_invert}|img2img_batch_files={img2img_batch_files}|img2img_batch_input_dir={img2img_batch_input_dir}|img2img_batch_output_dir={img2img_batch_output_dir}|img2img_batch_inpaint_mask_dir={img2img_batch_inpaint_mask_dir}|override_settings_texts={override_settings_texts}')
if mode == 5:
if img2img_batch_files is None or len(img2img_batch_files) == 0:
@@ -225,7 +225,7 @@ def img2img(id_task: str, state: str, mode: int,
width=width,
height=height,
full_quality=full_quality,
- restore_faces=restore_faces,
+ detailer=detailer,
tiling=tiling,
hidiffusion=hidiffusion,
init_images=[image],
diff --git a/modules/ipadapter.py b/modules/ipadapter.py
index de8522195..a92a30459 100644
--- a/modules/ipadapter.py
+++ b/modules/ipadapter.py
@@ -83,7 +83,7 @@ def crop_images(images, crops):
try:
for i in range(len(images)):
if crops[i]:
- from scripts.face_details import yolo # pylint: disable=no-name-in-module
+ from scripts.detailer import yolo # pylint: disable=no-name-in-module
yolo.load()
cropped = []
for image in images[i]:
diff --git a/modules/paths.py b/modules/paths.py
index 52bec2d0f..6f864da7f 100644
--- a/modules/paths.py
+++ b/modules/paths.py
@@ -123,6 +123,7 @@ def create_paths(opts):
create_path(fix_path('outdir_save'))
create_path(fix_path('outdir_video'))
create_path(fix_path('styles_dir'))
+ create_path(fix_path('yolo_dir'))
create_path(fix_path('wildcards_dir'))
diff --git a/modules/postprocess/codeformer_model.py b/modules/postprocess/codeformer_model.py
index e661335f9..4493ab8a3 100644
--- a/modules/postprocess/codeformer_model.py
+++ b/modules/postprocess/codeformer_model.py
@@ -1,7 +1,7 @@
import os
import cv2
import torch
-import modules.face_restoration
+import modules.detailer
from modules import shared, devices, modelloader, errors
from modules.paths import models_path
@@ -27,7 +27,7 @@ def setup_model(dirname):
return
try:
- class FaceRestorerCodeFormer(modules.face_restoration.FaceRestoration):
+ class FaceRestorerCodeFormer(modules.detailer.Detailer):
def name(self):
return "CodeFormer"
@@ -38,7 +38,7 @@ def setup_model(dirname):
def create_models(self):
from modules.postprocess.codeformer_arch import CodeFormer
- from facelib.utils.face_restoration_helper import FaceRestoreHelper
+ from facelib.utils.detailer_helper import FaceRestoreHelper
from facelib.detection.retinaface import retinaface
if self.net is not None and self.face_helper is not None:
self.net.to(devices.device)
@@ -100,7 +100,7 @@ def setup_model(dirname):
if original_resolution != restored_img.shape[0:2]:
restored_img = cv2.resize(restored_img, (0, 0), fx=original_resolution[1]/restored_img.shape[1], fy=original_resolution[0]/restored_img.shape[0], interpolation=cv2.INTER_LINEAR)
self.face_helper.clean_all()
- if shared.opts.face_restoration_unload:
+ if shared.opts.detailer_unload:
self.send_model_to(devices.cpu)
return restored_img
@@ -108,7 +108,7 @@ def setup_model(dirname):
have_codeformer = True
global codeformer # pylint: disable=global-statement
codeformer = FaceRestorerCodeFormer(dirname)
- shared.face_restorers.append(codeformer)
+ shared.detailers.append(codeformer)
except Exception as e:
errors.display(e, 'codeformer')
diff --git a/modules/postprocess/gfpgan_model.py b/modules/postprocess/gfpgan_model.py
index f17c91b2e..03274b7f4 100644
--- a/modules/postprocess/gfpgan_model.py
+++ b/modules/postprocess/gfpgan_model.py
@@ -56,7 +56,7 @@ def gfpgan_fix_faces(np_image):
model.face_helper.clean_all()
- if shared.opts.face_restoration_unload:
+ if shared.opts.detailer_unload:
send_model_to(model, devices.cpu)
return np_image
@@ -76,7 +76,7 @@ def setup_model(dirname):
install('gfpgan', quiet=True)
import gfpgan
import facexlib
- import modules.face_restoration
+ import modules.detailer
global user_path # pylint: disable=global-statement
global have_gfpgan # pylint: disable=global-statement
@@ -101,13 +101,13 @@ def setup_model(dirname):
have_gfpgan = True
gfpgan_constructor = gfpgan.GFPGANer
- class FaceRestorerGFPGAN(modules.face_restoration.FaceRestoration):
+ class FaceRestorerGFPGAN(modules.detailer.Detailer):
def name(self):
return "GFPGAN"
def restore(self, np_image, p=None): # pylint: disable=unused-argument
return gfpgan_fix_faces(np_image)
- shared.face_restorers.append(FaceRestorerGFPGAN())
+ shared.detailers.append(FaceRestorerGFPGAN())
except Exception as e:
errors.log.error(f'GFPGan failed to initialize: {e}')
diff --git a/modules/processing.py b/modules/processing.py
index c26668763..e0b89ef47 100644
--- a/modules/processing.py
+++ b/modules/processing.py
@@ -4,7 +4,7 @@ import time
from contextlib import nullcontext
import numpy as np
from PIL import Image, ImageOps
-from modules import shared, devices, errors, images, scripts, memstats, lowvram, script_callbacks, extra_networks, face_restoration, sd_hijack_freeu, sd_models, sd_vae, processing_helpers, timer
+from modules import shared, devices, errors, images, scripts, memstats, lowvram, script_callbacks, extra_networks, detailer, sd_hijack_freeu, sd_models, sd_vae, processing_helpers, timer
from modules.sd_hijack_hypertile import context_hypertile_vae, context_hypertile_unet
from modules.processing_class import StableDiffusionProcessing, StableDiffusionProcessingTxt2Img, StableDiffusionProcessingImg2Img, StableDiffusionProcessingControl # pylint: disable=unused-import
from modules.processing_info import create_infotext
@@ -50,8 +50,8 @@ class Processed:
self.image_cfg_scale = p.image_cfg_scale or 0
self.steps = p.steps or 0
self.batch_size = max(1, p.batch_size)
- self.restore_faces = p.restore_faces or False
- self.face_restoration_model = shared.opts.face_restoration_model if p.restore_faces else None
+ self.detailer = p.detailer or False
+ self.detailer_model = shared.opts.detailer_model if p.detailer else None
self.sd_model_hash = getattr(shared.sd_model, 'sd_model_hash', '') if model_data.sd_model is not None else ''
self.seed_resize_from_w = p.seed_resize_from_w
self.seed_resize_from_h = p.seed_resize_from_h
@@ -96,8 +96,8 @@ class Processed:
"cfg_scale": self.cfg_scale,
"steps": self.steps,
"batch_size": self.batch_size,
- "restore_faces": self.restore_faces,
- "face_restoration_model": self.face_restoration_model,
+ "detailer": self.detailer,
+ "detailer_model": self.detailer_model,
"sd_model_hash": self.sd_model_hash,
"seed_resize_from_w": self.seed_resize_from_w,
"seed_resize_from_h": self.seed_resize_from_h,
@@ -359,11 +359,11 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed:
else:
sample = validate_sample(sample)
image = Image.fromarray(sample)
- if p.restore_faces:
- if not p.do_not_save_samples and shared.opts.save_images_before_face_restoration:
+ if p.detailer:
+ if not p.do_not_save_samples and shared.opts.save_images_before_detailer:
images.save_image(Image.fromarray(sample), path=p.outpath_samples, basename="", seed=p.seeds[i], prompt=p.prompts[i], extension=shared.opts.samples_format, info=info, p=p, suffix="-before-face-restore")
p.ops.append('face')
- sample = face_restoration.restore_faces(sample, p)
+ sample = detailer.detail(sample, p)
if sample is not None:
image = Image.fromarray(sample)
if p.scripts is not None and isinstance(p.scripts, scripts.ScriptRunner):
@@ -439,8 +439,8 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed:
p.image_mask = ImageOps.invert(p.image_mask)
output_images.append(p.image_mask)
elif getattr(p, 'image_mask', None) is not None and isinstance(p.image_mask, Image.Image):
- if getattr(p, 'mask_for_facehires', None) is not None:
- output_images.append(p.mask_for_facehires)
+ if getattr(p, 'mask_for_detailer', None) is not None:
+ output_images.append(p.mask_for_detailer)
else:
output_images.append(p.image_mask)
diff --git a/modules/processing_class.py b/modules/processing_class.py
index dcf003f11..3eb35aa31 100644
--- a/modules/processing_class.py
+++ b/modules/processing_class.py
@@ -20,7 +20,7 @@ class StableDiffusionProcessing:
"""
The first set of paramaters: sd_models -> do_not_reload_embeddings represent the minimum required to create a StableDiffusionProcessing
"""
- def __init__(self, sd_model=None, outpath_samples=None, outpath_grids=None, prompt: str = "", styles: List[str] = None, seed: int = -1, subseed: int = -1, subseed_strength: float = 0, seed_resize_from_h: int = -1, seed_resize_from_w: int = -1, seed_enable_extras: bool = True, sampler_name: str = None, hr_sampler_name: str = None, batch_size: int = 1, n_iter: int = 1, steps: int = 50, cfg_scale: float = 7.0, image_cfg_scale: float = None, clip_skip: int = 1, width: int = 512, height: int = 512, full_quality: bool = True, restore_faces: bool = False, tiling: bool = False, hidiffusion: bool = False, do_not_save_samples: bool = False, do_not_save_grid: bool = False, extra_generation_params: Dict[Any, Any] = None, overlay_images: Any = None, negative_prompt: str = None, eta: float = None, do_not_reload_embeddings: bool = False, denoising_strength: float = 0, diffusers_guidance_rescale: float = 0.7, pag_scale: float = 0.0, pag_adaptive: float = 0.5, cfg_end: float = 1, resize_mode: int = 0, resize_name: str = 'None', resize_context: str = 'None', scale_by: float = 0, selected_scale_tab: int = 0, hdr_mode: int = 0, hdr_brightness: float = 0, hdr_color: float = 0, hdr_sharpen: float = 0, hdr_clamp: bool = False, hdr_boundary: float = 4.0, hdr_threshold: float = 0.95, hdr_maximize: bool = False, hdr_max_center: float = 0.6, hdr_max_boundry: float = 1.0, hdr_color_picker: str = None, hdr_tint_ratio: float = 0, override_settings: Dict[str, Any] = None, override_settings_restore_afterwards: bool = True, sampler_index: int = None, script_args: list = None): # pylint: disable=unused-argument
+ def __init__(self, sd_model=None, outpath_samples=None, outpath_grids=None, prompt: str = "", styles: List[str] = None, seed: int = -1, subseed: int = -1, subseed_strength: float = 0, seed_resize_from_h: int = -1, seed_resize_from_w: int = -1, seed_enable_extras: bool = True, sampler_name: str = None, hr_sampler_name: str = None, batch_size: int = 1, n_iter: int = 1, steps: int = 50, cfg_scale: float = 7.0, image_cfg_scale: float = None, clip_skip: int = 1, width: int = 512, height: int = 512, full_quality: bool = True, detailer: bool = False, tiling: bool = False, hidiffusion: bool = False, do_not_save_samples: bool = False, do_not_save_grid: bool = False, extra_generation_params: Dict[Any, Any] = None, overlay_images: Any = None, negative_prompt: str = None, eta: float = None, do_not_reload_embeddings: bool = False, denoising_strength: float = 0, diffusers_guidance_rescale: float = 0.7, pag_scale: float = 0.0, pag_adaptive: float = 0.5, cfg_end: float = 1, resize_mode: int = 0, resize_name: str = 'None', resize_context: str = 'None', scale_by: float = 0, selected_scale_tab: int = 0, hdr_mode: int = 0, hdr_brightness: float = 0, hdr_color: float = 0, hdr_sharpen: float = 0, hdr_clamp: bool = False, hdr_boundary: float = 4.0, hdr_threshold: float = 0.95, hdr_maximize: bool = False, hdr_max_center: float = 0.6, hdr_max_boundry: float = 1.0, hdr_color_picker: str = None, hdr_tint_ratio: float = 0, override_settings: Dict[str, Any] = None, override_settings_restore_afterwards: bool = True, sampler_index: int = None, script_args: list = None): # pylint: disable=unused-argument
self.state: str = ''
self.skip = []
self.outpath_samples: str = outpath_samples
@@ -50,7 +50,7 @@ class StableDiffusionProcessing:
self.width: int = width
self.height: int = height
self.full_quality: bool = full_quality
- self.restore_faces: bool = restore_faces
+ self.detailer: bool = detailer
self.tiling: bool = tiling
self.hidiffusion: bool = hidiffusion
self.do_not_save_samples: bool = do_not_save_samples
diff --git a/modules/processing_diffusers.py b/modules/processing_diffusers.py
index 22293fb02..c06c3d3de 100644
--- a/modules/processing_diffusers.py
+++ b/modules/processing_diffusers.py
@@ -17,7 +17,7 @@ orig_pipeline = shared.sd_model
def restore_state(p: processing.StableDiffusionProcessing):
- if p.state in ['reprocess_refine', 'reprocess_face']:
+ if p.state in ['reprocess_refine', 'reprocess_detail']:
# validate
if last_p is None:
shared.log.warning(f'Restore state: op={p.state} last state missing')
@@ -43,9 +43,9 @@ def restore_state(p: processing.StableDiffusionProcessing):
p.hr_scale, p.hr_upscaler, p.hr_resize_mode, p.hr_resize_context, p.hr_resize_x, p.hr_resize_y, p.hr_upscale_to_x, p.hr_upscale_to_y = hr_scale, hr_upscaler, hr_resize_mode, hr_resize_context, hr_resize_x, hr_resize_y, hr_upscale_to_x, hr_upscale_to_y
p.height, p.width, p.scale_by, p.resize_mode, p.resize_name, p.resize_context = height, width, scale_by, resize_mode, resize_name, resize_context
p.init_images = None
- if state == 'reprocess_face':
+ if state == 'reprocess_detail':
p.skip = ['encode', 'base', 'hires']
- p.restore_faces = True
+ p.detailer = True
shared.log.info(f'Restore state: op={p.state} skip={p.skip}')
return p
diff --git a/modules/processing_info.py b/modules/processing_info.py
index ad0a455fa..b0f4f20d8 100644
--- a/modules/processing_info.py
+++ b/modules/processing_info.py
@@ -124,7 +124,7 @@ def create_infotext(p: StableDiffusionProcessing, all_prompts=None, all_seeds=No
args['Size scale mask'] = p.scale_by_mask
args['Size name mask'] = p.resize_name_mask
if 'face' in p.ops:
- args["Face restoration"] = shared.opts.face_restoration_model
+ args["Face restoration"] = shared.opts.detailer_model
if 'color' in p.ops:
args["Color correction"] = True
# embeddings
diff --git a/modules/processing_original.py b/modules/processing_original.py
index bd6a8b466..852eb9a37 100644
--- a/modules/processing_original.py
+++ b/modules/processing_original.py
@@ -90,11 +90,11 @@ def sample_txt2img(p: processing.StableDiffusionProcessingTxt2Img, conditioning,
for i, x_sample in enumerate(decoded_samples):
x_sample = validate_sample(x_sample)
image = Image.fromarray(x_sample)
- bak_extra_generation_params, bak_restore_faces = p.extra_generation_params, p.restore_faces
+ bak_extra_generation_params, bak_detailer = p.extra_generation_params, p.detailer
p.extra_generation_params = {}
- p.restore_faces = False
+ p.detailer = False
info = processing.create_infotext(p, p.all_prompts, p.all_seeds, p.all_subseeds, [], iteration=p.iteration, position_in_batch=i)
- p.extra_generation_params, p.restore_faces = bak_extra_generation_params, bak_restore_faces
+ p.extra_generation_params, p.detailer = bak_extra_generation_params, bak_detailer
images.save_image(image, p.outpath_samples, "", seeds[i], prompts[i], shared.opts.samples_format, info=info, suffix="-before-hires")
if latent_scale_mode is None or p.hr_force: # non-latent upscaling
shared.state.job = 'Upscale'
diff --git a/modules/sd_samplers.py b/modules/sd_samplers.py
index cf72301e4..0ea6dbdee 100644
--- a/modules/sd_samplers.py
+++ b/modules/sd_samplers.py
@@ -54,7 +54,7 @@ def create_sampler(name, model):
model.prior_pipe.scheduler = copy.deepcopy(model.default_scheduler)
model.prior_pipe.scheduler.config.clip_sample = False
config = {k: v for k, v in model.scheduler.config.items() if not k.startswith('_')}
- shared.log.debug(f'Sampler default {type(model.scheduler).__name__}: {config}')
+ shared.log.debug(f'Sampler: sampler=default class={model.scheduler.__class__.__name__}: {config}')
return model.scheduler
config = find_sampler_config(name)
if config is None or config.constructor is None:
diff --git a/modules/shared.py b/modules/shared.py
index abcfbdbab..aba93bbaf 100644
--- a/modules/shared.py
+++ b/modules/shared.py
@@ -43,7 +43,8 @@ locking_available = True
clip_model = None
interrogator = modules.interrogate.InterrogateModels(os.path.join("models", "interrogate"))
sd_upscalers = []
-face_restorers = []
+detailers = []
+yolo = None
tab_names = []
extra_networks = []
options_templates = {}
@@ -592,6 +593,7 @@ options_templates.update(options_section(('system-paths', "System Paths"), {
"embeddings_dir": OptionInfo(os.path.join(paths.models_path, 'embeddings'), "Folder with textual inversion embeddings", folder=True),
"hypernetwork_dir": OptionInfo(os.path.join(paths.models_path, 'hypernetworks'), "Folder with Hypernetwork models", folder=True),
"control_dir": OptionInfo(os.path.join(paths.models_path, 'control'), "Folder with Control models", folder=True),
+ "yolo_dir": OptionInfo(os.path.join(paths.models_path, 'yolo'), "Folder with Yolo models", folder=True),
"codeformer_models_path": OptionInfo(os.path.join(paths.models_path, 'Codeformer'), "Folder with codeformer models", folder=True),
"gfpgan_models_path": OptionInfo(os.path.join(paths.models_path, 'GFPGAN'), "Folder with GFPGAN models", folder=True),
"esrgan_models_path": OptionInfo(os.path.join(paths.models_path, 'ESRGAN'), "Folder with ESRGAN models", folder=True),
@@ -644,7 +646,7 @@ options_templates.update(options_section(('saving-images', "Image Options"), {
"save_init_img": OptionInfo(False, "Save init images"),
"save_images_before_highres_fix": OptionInfo(False, "Save image before hires"),
"save_images_before_refiner": OptionInfo(False, "Save image before refiner"),
- "save_images_before_face_restoration": OptionInfo(False, "Save image before face restoration"),
+ "save_images_before_detailer": OptionInfo(False, "Save image before detailer"),
"save_images_before_color_correction": OptionInfo(False, "Save image before color correction"),
"save_mask": OptionInfo(False, "Save inpainting mask"),
"save_mask_composite": OptionInfo(False, "Save inpainting masked composite"),
@@ -782,18 +784,19 @@ options_templates.update(options_section(('postprocessing', "Postprocessing"), {
"img2img_extra_noise": OptionInfo(0.0, "Extra noise multiplier for img2img", gr.Slider, {"minimum": 0.0, "maximum": 1.0, "step": 0.01}),
"CLIP_stop_at_last_layers": OptionInfo(1, "Clip skip", gr.Slider, {"minimum": 1, "maximum": 8, "step": 1, "visible": False}),
- "postprocessing_sep_face_restoration": OptionInfo("Face Restoration
", "", gr.HTML),
- "face_restoration_model": OptionInfo("Face HiRes", "Face restoration model", gr.Radio, lambda: {"choices": [x.name() for x in face_restorers]}),
- "facehires_sep": OptionInfo("Face restore
", "", gr.HTML),
- "facehires_conf": OptionInfo(0.6, "Min confidence", gr.Slider, {"minimum": 0.0, "maximum": 1.0, "step": 0.05}),
- "facehires_max": OptionInfo(5, "Max faces", gr.Slider, {"minimum": 1, "maximum": 10, "step": 1}),
- "facehires_iou": OptionInfo(0.5, "Max face overlap", gr.Slider, {"minimum": 0, "maximum": 1.0, "step": 0.05}),
- "facehires_min_size": OptionInfo(0, "Min face size", gr.Slider, {"minimum": 0, "maximum": 1024, "step": 1}),
- "facehires_max_size": OptionInfo(0, "Max face size", gr.Slider, {"minimum": 0, "maximum": 1024, "step": 1}),
- "facehires_padding": OptionInfo(20, "Face padding", gr.Slider, {"minimum": 0, "maximum": 100, "step": 1}),
- "facehires_strength": OptionInfo(0.0, "Face restore strength", gr.Slider, {"minimum": 0, "maximum": 1, "step": 0.01}),
+ "postprocessing_sep_detailer": OptionInfo("Detailer
", "", gr.HTML),
+ "detailer_model": OptionInfo("Detailer", "Detailer model", gr.Radio, lambda: {"choices": [x.name() for x in detailers]}),
+ "detailer_sep": OptionInfo("Detailer
", "", gr.HTML),
+ "detailer_conf": OptionInfo(0.6, "Min confidence", gr.Slider, {"minimum": 0.0, "maximum": 1.0, "step": 0.05}),
+ "detailer_max": OptionInfo(5, "Max detected", gr.Slider, {"minimum": 1, "maximum": 10, "step": 1}),
+ "detailer_iou": OptionInfo(0.5, "Max overlap", gr.Slider, {"minimum": 0, "maximum": 1.0, "step": 0.05}),
+ "detailer_min_size": OptionInfo(0, "Min object size", gr.Slider, {"minimum": 0, "maximum": 1024, "step": 1}),
+ "detailer_max_size": OptionInfo(0, "Max object size", gr.Slider, {"minimum": 0, "maximum": 1024, "step": 1}),
+ "detailer_padding": OptionInfo(20, "Object padding", gr.Slider, {"minimum": 0, "maximum": 100, "step": 1}),
+ "detailer_strength": OptionInfo(0.0, "Detailer strength", gr.Slider, {"minimum": 0, "maximum": 1, "step": 0.01}),
+ "detailer_models": OptionInfo(['Face yolo-8n'], "Detailer models", gr.Dropdown, lambda: {"multiselect":True, "choices": list(yolo.list)}),
"code_former_weight": OptionInfo(0.2, "CodeFormer weight parameter", gr.Slider, {"minimum": 0, "maximum": 1, "step": 0.01}),
- "face_restoration_unload": OptionInfo(False, "Move model to CPU when complete"),
+ "detailer_unload": OptionInfo(False, "Move detailer model to CPU when complete"),
"postprocessing_sep_upscalers": OptionInfo("Upscaling
", "", gr.HTML),
"upscaler_unload": OptionInfo(False, "Unload upscaler after processing"),
diff --git a/modules/txt2img.py b/modules/txt2img.py
index e23b3efd7..077a3db9e 100644
--- a/modules/txt2img.py
+++ b/modules/txt2img.py
@@ -11,7 +11,7 @@ debug('Trace: PROCESS')
def txt2img(id_task, state,
prompt, negative_prompt, prompt_styles,
steps, sampler_index, hr_sampler_index,
- full_quality, restore_faces, tiling, hidiffusion,
+ full_quality, detailer, tiling, hidiffusion,
n_iter, batch_size,
cfg_scale, image_cfg_scale, diffusers_guidance_rescale, pag_scale, pag_adaptive, cfg_end,
clip_skip,
@@ -24,7 +24,7 @@ def txt2img(id_task, state,
override_settings_texts,
*args):
- debug(f'txt2img: id_task={id_task}|prompt={prompt}|negative={negative_prompt}|styles={prompt_styles}|steps={steps}|sampler_index={sampler_index}|hr_sampler_index={hr_sampler_index}|full_quality={full_quality}|restore_faces={restore_faces}|tiling={tiling}|hidiffusion={hidiffusion}|batch_count={n_iter}|batch_size={batch_size}|cfg_scale={cfg_scale}|clip_skip={clip_skip}|seed={seed}|subseed={subseed}|subseed_strength={subseed_strength}|seed_resize_from_h={seed_resize_from_h}|seed_resize_from_w={seed_resize_from_w}|height={height}|width={width}|enable_hr={enable_hr}|denoising_strength={denoising_strength}|hr_resize_mode={hr_resize_mode}|hr_resize_context={hr_resize_context}|hr_scale={hr_scale}|hr_upscaler={hr_upscaler}|hr_force={hr_force}|hr_second_pass_steps={hr_second_pass_steps}|hr_resize_x={hr_resize_x}|hr_resize_y={hr_resize_y}|image_cfg_scale={image_cfg_scale}|diffusers_guidance_rescale={diffusers_guidance_rescale}|refiner_steps={refiner_steps}|refiner_start={refiner_start}|refiner_prompt={refiner_prompt}|refiner_negative={refiner_negative}|override_settings={override_settings_texts}')
+ debug(f'txt2img: id_task={id_task}|prompt={prompt}|negative={negative_prompt}|styles={prompt_styles}|steps={steps}|sampler_index={sampler_index}|hr_sampler_index={hr_sampler_index}|full_quality={full_quality}|detailer={detailer}|tiling={tiling}|hidiffusion={hidiffusion}|batch_count={n_iter}|batch_size={batch_size}|cfg_scale={cfg_scale}|clip_skip={clip_skip}|seed={seed}|subseed={subseed}|subseed_strength={subseed_strength}|seed_resize_from_h={seed_resize_from_h}|seed_resize_from_w={seed_resize_from_w}|height={height}|width={width}|enable_hr={enable_hr}|denoising_strength={denoising_strength}|hr_resize_mode={hr_resize_mode}|hr_resize_context={hr_resize_context}|hr_scale={hr_scale}|hr_upscaler={hr_upscaler}|hr_force={hr_force}|hr_second_pass_steps={hr_second_pass_steps}|hr_resize_x={hr_resize_x}|hr_resize_y={hr_resize_y}|image_cfg_scale={image_cfg_scale}|diffusers_guidance_rescale={diffusers_guidance_rescale}|refiner_steps={refiner_steps}|refiner_start={refiner_start}|refiner_prompt={refiner_prompt}|refiner_negative={refiner_negative}|override_settings={override_settings_texts}')
if shared.sd_model is None:
shared.log.warning('Model not loaded')
@@ -65,7 +65,7 @@ def txt2img(id_task, state,
width=width,
height=height,
full_quality=full_quality,
- restore_faces=restore_faces,
+ detailer=detailer,
tiling=tiling,
hidiffusion=hidiffusion,
enable_hr=enable_hr,
diff --git a/modules/ui_control.py b/modules/ui_control.py
index a65a6ef8f..cb7ae2921 100644
--- a/modules/ui_control.py
+++ b/modules/ui_control.py
@@ -102,7 +102,7 @@ def create_ui(_blocks: gr.Blocks=None):
with gr.Row(elem_id='control_settings'):
- full_quality, restore_faces, tiling, hidiffusion = ui_sections.create_options('control')
+ full_quality, detailer, tiling, hidiffusion = ui_sections.create_options('control')
state = gr.Textbox(value='', visible=False)
with gr.Accordion(open=False, label="Input", elem_id="control_input", elem_classes=["small-accordion"]):
@@ -534,7 +534,7 @@ def create_ui(_blocks: gr.Blocks=None):
prompt, negative, styles,
steps, sampler_index,
seed, subseed, subseed_strength, seed_resize_from_h, seed_resize_from_w,
- cfg_scale, clip_skip, image_cfg_scale, diffusers_guidance_rescale, pag_scale, pag_adaptive, cfg_end, full_quality, restore_faces, tiling, hidiffusion,
+ cfg_scale, clip_skip, image_cfg_scale, diffusers_guidance_rescale, pag_scale, pag_adaptive, cfg_end, full_quality, detailer, tiling, hidiffusion,
hdr_mode, hdr_brightness, hdr_color, hdr_sharpen, hdr_clamp, hdr_boundary, hdr_threshold, hdr_maximize, hdr_max_center, hdr_max_boundry, hdr_color_picker, hdr_tint_ratio,
resize_mode_before, resize_name_before, resize_context_before, width_before, height_before, scale_by_before, selected_scale_tab_before,
resize_mode_after, resize_name_after, resize_context_after, width_after, height_after, scale_by_after, selected_scale_tab_after,
@@ -616,7 +616,7 @@ def create_ui(_blocks: gr.Blocks=None):
(image_cfg_scale, "Image CFG scale"),
(diffusers_guidance_rescale, "CFG rescale"),
(full_quality, "Full quality"),
- (restore_faces, "Face restoration"),
+ (detailer, "Face restoration"),
(tiling, "Tiling"),
(hidiffusion, "HiDiffusion"),
# second pass
diff --git a/modules/ui_img2img.py b/modules/ui_img2img.py
index 1cbf30345..ec37f5e23 100644
--- a/modules/ui_img2img.py
+++ b/modules/ui_img2img.py
@@ -130,7 +130,7 @@ def create_ui():
refiner_start = gr.Slider(minimum=0.0, maximum=1.0, step=0.05, label='Denoise start', value=0.0, elem_id="img2img_refiner_start")
cfg_scale, clip_skip, image_cfg_scale, diffusers_guidance_rescale, pag_scale, pag_adaptive, cfg_end = ui_sections.create_advanced_inputs('img2img')
- full_quality, restore_faces, tiling, hidiffusion = ui_sections.create_options('img2img')
+ full_quality, detailer, tiling, hidiffusion = ui_sections.create_options('img2img')
hdr_mode, hdr_brightness, hdr_color, hdr_sharpen, hdr_clamp, hdr_boundary, hdr_threshold, hdr_maximize, hdr_max_center, hdr_max_boundry, hdr_color_picker, hdr_tint_ratio = ui_sections.create_correction_inputs('img2img')
# with gr.Group(elem_id="inpaint_controls", visible=False) as inpaint_controls:
@@ -177,7 +177,7 @@ def create_ui():
sampler_index,
mask_blur, mask_alpha,
inpainting_fill,
- full_quality, restore_faces, tiling, hidiffusion,
+ full_quality, detailer, tiling, hidiffusion,
batch_count, batch_size,
cfg_scale, image_cfg_scale,
diffusers_guidance_rescale, pag_scale, pag_adaptive, cfg_end,
@@ -267,7 +267,7 @@ def create_ui():
(clip_skip, "Clip skip"),
(diffusers_guidance_rescale, "CFG rescale"),
(full_quality, "Full quality"),
- (restore_faces, "Face restoration"),
+ (detailer, "Detailer"),
(tiling, "Tiling"),
(hidiffusion, "HiDiffusion"),
# inpaint
diff --git a/modules/ui_sections.py b/modules/ui_sections.py
index c30605cfd..aa076b1ea 100644
--- a/modules/ui_sections.py
+++ b/modules/ui_sections.py
@@ -32,7 +32,7 @@ def create_toprow(is_img2img: bool = False, id_part: str = None):
reprocess.append(gr.Button('Reprocess', elem_id=f"{id_part}_reprocess", variant='primary', visible=True))
reprocess.append(gr.Button('Reprocess decode', elem_id=f"{id_part}_reprocess_decode", variant='primary', visible=False))
reprocess.append(gr.Button('Reprocess refine', elem_id=f"{id_part}_reprocess_refine", variant='primary', visible=False))
- reprocess.append(gr.Button('Reprocess face', elem_id=f"{id_part}_reprocess_face", variant='primary', visible=False))
+ reprocess.append(gr.Button('Reprocess face', elem_id=f"{id_part}_reprocess_detail", variant='primary', visible=False))
with gr.Row(elem_id=f"{id_part}_generate_line2"):
interrupt = gr.Button('Stop', elem_id=f"{id_part}_interrupt")
interrupt.click(fn=lambda: shared.state.interrupt(), _js="requestInterrupt", inputs=[], outputs=[])
@@ -149,10 +149,10 @@ def create_seed_inputs(tab, reuse_visible=True):
def create_options(tab):
with gr.Row(elem_id=f"{tab}_advanced_options"):
full_quality = gr.Checkbox(label='Full quality', value=True, elem_id=f"{tab}_full_quality")
- restore_faces = gr.Checkbox(label='Face restore', value=False, elem_id=f"{tab}_restore_faces")
+ detailer = gr.Checkbox(label='Detailer', value=False, elem_id=f"{tab}_detailer")
tiling = gr.Checkbox(label='Tiling', value=False, elem_id=f"{tab}_tiling")
hidiffusion = gr.Checkbox(label='HiDiffusion', value=False, elem_id=f"{tab}_hidiffusion")
- return full_quality, restore_faces, tiling, hidiffusion
+ return full_quality, detailer, tiling, hidiffusion
def create_cfg_inputs(tab):
diff --git a/modules/ui_txt2img.py b/modules/ui_txt2img.py
index f4ecf3e49..171bc9ac1 100644
--- a/modules/ui_txt2img.py
+++ b/modules/ui_txt2img.py
@@ -39,7 +39,7 @@ def create_ui():
batch_count, batch_size = ui_sections.create_batch_inputs('txt2img', accordion=False)
cfg_scale, cfg_end = ui_sections.create_cfg_inputs('txt2img')
steps, sampler_index = ui_sections.create_sampler_and_steps_selection(None, "txt2img")
- full_quality, restore_faces, tiling, hidiffusion = ui_sections.create_options('txt2img')
+ full_quality, detailer, tiling, hidiffusion = ui_sections.create_options('txt2img')
with gr.Group(elem_classes="settings-accordion"):
with gr.Accordion(open=False, label="Samplers", elem_classes=["small-accordion"], elem_id="txt2img_sampler_group"):
@@ -64,7 +64,7 @@ def create_ui():
dummy_component, state,
txt2img_prompt, txt2img_negative_prompt, txt2img_prompt_styles,
steps, sampler_index, hr_sampler_index,
- full_quality, restore_faces, tiling, hidiffusion,
+ full_quality, detailer, tiling, hidiffusion,
batch_count, batch_size,
cfg_scale, image_cfg_scale, diffusers_guidance_rescale, pag_scale, pag_adaptive, cfg_end,
clip_skip,
@@ -120,7 +120,7 @@ def create_ui():
(image_cfg_scale, "Image CFG scale"),
(diffusers_guidance_rescale, "CFG rescale"),
(full_quality, "Full quality"),
- (restore_faces, "Face restoration"),
+ (detailer, "Detailer"),
(tiling, "Tiling"),
(hidiffusion, "HiDiffusion"),
# second pass
diff --git a/scripts/face_details.py b/scripts/detailer.py
similarity index 58%
rename from scripts/face_details.py
rename to scripts/detailer.py
index 0d803ccb3..e9f94667d 100644
--- a/scripts/face_details.py
+++ b/scripts/detailer.py
@@ -2,34 +2,49 @@ import os
import numpy as np
from PIL import Image, ImageDraw
from modules import shared, processing
-from modules.face_restoration import FaceRestoration
+from modules.detailer import Detailer
from modules import devices, processing_class
-class YoLoResult:
- def __init__(self, score: float, box: list[int], mask: Image.Image = None, face: Image.Image = None, size: float = 0, width = 0, height = 0, args = {}):
+PREDEFINED = { #
+ 'Face yolo-8n': 'https://huggingface.co/vladmandic/yolo-detailers/resolve/main/face-yolo8n.pt',
+ 'Eyefull paired v2': 'https://huggingface.co/vladmandic/yolo-detailers/resolve/main/eyeful-paired-v2.pt',
+}
+
+
+class YoloResult:
+ def __init__(self, score: float, box: list[int], mask: Image.Image = None, item: Image.Image = None, size: float = 0, width = 0, height = 0, args = {}):
self.score = score
self.box = box
self.mask = mask
- self.face = face
+ self.item = item
self.size = size
self.width = width
self.height = height
self.args = args
-class FaceRestorerYolo(FaceRestoration):
- def name(self):
- return "Face HiRes"
-
+class YoloRestorer(Detailer):
def __init__(self):
- from modules import paths
- self.model = None
- self.model_dir = os.path.join(paths.models_path, 'yolo')
- self.model_name = 'yolov8n-face.pt'
- self.model_url = 'https://github.com/akanametov/yolov8-face/releases/download/v0.0.0/yolov8n-face.pt'
- # self.model_name = 'yolov9-c-face.pt'
- # self.model_url = 'https://github.com/akanametov/yolov9-face/releases/download/1.0/yolov9-c-face.pt'
+ super().__init__()
+ self.models = {}
+ self.list = {}
+ self.enumerate()
+
+ def name(self):
+ return "Detailer"
+
+ def enumerate(self):
+ self.list.clear()
+ files = []
+ for k, v in PREDEFINED.items():
+ self.list[k] = v
+ files.append(os.path.basename(v))
+ for f in os.listdir(shared.opts.yolo_dir):
+ if f not in files:
+ name = os.path.basename(f)
+ self.list[name] = os.path.join(shared.opts.yolo_dir, f)
+ shared.log.info(f'Available Yolo: path="{shared.opts.yolo_dir} items={len(list(self.list))}')
def dependencies(self):
import installer
@@ -37,6 +52,7 @@ class FaceRestorerYolo(FaceRestoration):
def predict(
self,
+ model,
image: Image.Image,
imgsz: int = 640,
half: bool = True,
@@ -45,16 +61,16 @@ class FaceRestorerYolo(FaceRestoration):
agnostic: bool = False,
retina: bool = False,
mask: bool = True,
- offload: bool = shared.opts.face_restoration_unload,
- ) -> list[YoLoResult]:
+ offload: bool = shared.opts.detailer_unload,
+ ) -> list[YoloResult]:
args = {
- 'conf': shared.opts.facehires_conf,
- 'iou': shared.opts.facehires_iou,
- 'max_det': shared.opts.facehires_max,
+ 'conf': shared.opts.detailer_conf,
+ 'iou': shared.opts.detailer_iou,
+ 'max_det': shared.opts.detailer_max,
}
- self.model.to(device)
- predictions = self.model.predict(
+ model.to(device)
+ predictions = model.predict(
source=[image],
stream=False,
verbose=False,
@@ -67,7 +83,8 @@ class FaceRestorerYolo(FaceRestoration):
**args
)
if offload:
- self.model.to('cpu')
+ model.to('cpu')
+
result = []
for prediction in predictions:
boxes = prediction.boxes.xyxy.detach().int().cpu().numpy() if prediction.boxes is not None else []
@@ -77,41 +94,50 @@ class FaceRestorerYolo(FaceRestoration):
mask_image = None
w, h = box[2] - box[0], box[3] - box[1]
size = w * h / (image.width * image.height)
- if (min(w, h) > shared.opts.facehires_min_size if shared.opts.facehires_min_size > 0 else True) and (max(w, h) < shared.opts.facehires_max_size if shared.opts.facehires_max_size > 0 else True):
+ if (min(w, h) > shared.opts.detailer_min_size if shared.opts.detailer_min_size > 0 else True) and (max(w, h) < shared.opts.detailer_max_size if shared.opts.detailer_max_size > 0 else True):
if mask:
mask_image = image.copy()
mask_image = Image.new('L', image.size, 0)
draw = ImageDraw.Draw(mask_image)
draw.rectangle(box, fill="white", outline=None, width=0)
- face_image = image.crop(box)
- result.append(YoLoResult(score=round(score, 2), box=box, mask=mask_image, face=face_image, size=size, width=w, height=h, args=args))
+ cropped = image.crop(box)
+ result.append(YoloResult(score=round(score, 2), box=box, mask=mask_image, item=cropped, size=size, width=w, height=h, args=args))
return result
- def load(self):
+ def load(self, model_name: str = None):
from modules import modelloader
self.dependencies()
- if self.model is None:
- model_file = modelloader.load_file_from_url(url=self.model_url, model_dir=self.model_dir, file_name=self.model_name)
+ if model_name is None:
+ model_name = list(self.list)[0]
+ if model_name in self.models:
+ return model_name, self.models[model_name]
+ else:
+ model_url = self.list.get(model_name)
+ file_name = os.path.basename(model_url)
+ model_file = modelloader.load_file_from_url(url=model_url, model_dir=shared.opts.yolo_dir, file_name=file_name)
if model_file is not None:
- shared.log.info(f'Load: type=FaceHires model={model_file}')
+ shared.log.info(f'Load: type=Detailer name="{model_name}" model="{model_file}"')
from ultralytics import YOLO # pylint: disable=import-outside-toplevel
- self.model = YOLO(model_file)
+ model = YOLO(model_file)
+ self.models[model_name] = model
+ return model_name, model
+ return None
def restore(self, np_image, p: processing.StableDiffusionProcessing = None):
if hasattr(p, 'recursion'):
return
- if not hasattr(p, 'facehires'):
- p.facehires = 0
- if np_image is None or p.facehires >= p.batch_size * p.n_iter:
+ if not hasattr(p, 'detailer_active'):
+ p.detailer_active = 0
+ if np_image is None or p.detailer_active >= p.batch_size * p.n_iter:
return np_image
- self.load()
- if self.model is None:
- shared.log.debug('Face HiRes: model not loaded')
+ name, model = self.load()
+ if model is None:
+ shared.log.warning(f'Detailer: model="{name}" not loaded')
return np_image
image = Image.fromarray(np_image)
- faces = self.predict(image)
- if len(faces) == 0:
- shared.log.debug('Face HiRes: no faces detected')
+ items = self.predict(model, image)
+ if len(items) == 0:
+ shared.log.info(f'Detailer: model="{name}" no items detected')
return np_image
# create backups
@@ -131,17 +157,17 @@ class FaceRestorerYolo(FaceRestoration):
'sampler_name': orig_p.get('hr_sampler_name', 'default'),
'steps': orig_p.get('hr_second_pass_steps', 0),
'negative_prompt': orig_p.get('refiner_negative', ''),
- 'denoising_strength': shared.opts.facehires_strength if shared.opts.facehires_strength > 0 else orig_p.get('denoising_strength', 0.3),
+ 'denoising_strength': shared.opts.detailer_strength if shared.opts.detailer_strength > 0 else orig_p.get('denoising_strength', 0.3),
'styles': [],
'prompt': orig_p.get('refiner_prompt', ''),
'mask_blur': 10,
- 'inpaint_full_res_padding': shared.opts.facehires_padding,
- 'restore_faces': True,
+ 'inpaint_full_res_padding': shared.opts.detailer_padding,
+ 'detailer': True,
'width': resolution,
'height': resolution,
}
if args['denoising_strength'] == 0:
- shared.log.debug('Face HiRes skip: strength=0')
+ shared.log.debug(f'Detailer: model="{name}" strength=0 skip')
return np_image
control_pipeline = None
orig_class = shared.sd_model.__class__
@@ -151,7 +177,7 @@ class FaceRestorerYolo(FaceRestoration):
run.restore_pipeline()
p = processing_class.switch_class(p, processing.StableDiffusionProcessingImg2Img, args)
- p.facehires += 1 # set flag to avoid recursion
+ p.detailer_active += 1 # set flag to avoid recursion
if p.steps < 1:
p.steps = orig_p.get('steps', 0)
@@ -160,23 +186,23 @@ class FaceRestorerYolo(FaceRestoration):
if len(p.negative_prompt) == 0:
p.negative_prompt = orig_p.get('all_negative_prompts', [''])[0]
- report = [{'score': f.score, 'size': f'{f.width}x{f.height}' } for f in faces]
- shared.log.debug(f'Face HiRes: faces={report} args={faces[0].args} denoise={p.denoising_strength} blur={p.mask_blur} width={p.width} height={p.height} padding={p.inpaint_full_res_padding}')
+ report = [{'score': i.score, 'size': f'{i.width}x{i.height}' } for i in items]
+ shared.log.info(f'Detailer: model="{name}" items={report} args={items[0].args} denoise={p.denoising_strength} blur={p.mask_blur} width={p.width} height={p.height} padding={p.inpaint_full_res_padding}')
mask_all = []
p.state = ''
- for face in faces:
- if face.mask is None:
+ for item in items:
+ if item.mask is None:
continue
p.init_images = [image]
- p.image_mask = [face.mask]
- # mask_all.append(face.mask)
+ p.image_mask = [item.mask]
+ # mask_all.append(item.mask)
p.recursion = True
pp = processing.process_images_inner(p)
del p.recursion
p.overlay_images = None # skip applying overlay twice
if pp is not None and pp.images is not None and len(pp.images) > 0:
- image = pp.images[0] # update image to be reused for next face
+ image = pp.images[0] # update image to be reused for next item
if len(pp.images) > 1:
mask_all.append(pp.images[1])
@@ -197,10 +223,11 @@ class FaceRestorerYolo(FaceRestoration):
p.image_mask = blend([np.array(m) for m in mask_all])
# combined = blend([np_image, p.image_mask])
# combined = Image.fromarray(combined)
- # combined.save('/tmp/face.png')
+ # combined.save('/tmp/item.png')
p.image_mask = Image.fromarray(p.image_mask)
return np_image
-yolo = FaceRestorerYolo()
-shared.face_restorers.append(yolo)
+yolo = YoloRestorer()
+shared.detailers.append(yolo)
+shared.yolo = yolo
diff --git a/scripts/prompts_from_file.py b/scripts/prompts_from_file.py
index 465a44974..24dbc0b86 100644
--- a/scripts/prompts_from_file.py
+++ b/scripts/prompts_from_file.py
@@ -45,7 +45,7 @@ prompt_tags = {
"cfg_scale": process_float_tag,
"width": process_int_tag,
"height": process_int_tag,
- "restore_faces": process_boolean_tag,
+ "detailer": process_boolean_tag,
"tiling": process_boolean_tag,
"do_not_save_samples": process_boolean_tag,
"do_not_save_grid": process_boolean_tag
diff --git a/scripts/xyz_grid_classes.py b/scripts/xyz_grid_classes.py
index f92a4fd43..db5684f97 100644
--- a/scripts/xyz_grid_classes.py
+++ b/scripts/xyz_grid_classes.py
@@ -1,4 +1,4 @@
-from scripts.xyz_grid_shared import apply_field, apply_task_args, apply_setting, apply_prompt, apply_order, apply_sampler, apply_hr_sampler_name, confirm_samplers, apply_checkpoint, apply_refiner, apply_unet, apply_dict, apply_clip_skip, apply_vae, list_lora, apply_lora, apply_te, apply_styles, apply_upscaler, apply_context, apply_face_restore, apply_override, apply_processing, apply_options, apply_seed, format_value_add_label, format_value, format_value_join_list, do_nothing, format_nothing, str_permutations # pylint: disable=no-name-in-module
+from scripts.xyz_grid_shared import apply_field, apply_task_args, apply_setting, apply_prompt, apply_order, apply_sampler, apply_hr_sampler_name, confirm_samplers, apply_checkpoint, apply_refiner, apply_unet, apply_dict, apply_clip_skip, apply_vae, list_lora, apply_lora, apply_te, apply_styles, apply_upscaler, apply_context, apply_detailer, apply_override, apply_processing, apply_options, apply_seed, format_value_add_label, format_value, format_value_join_list, do_nothing, format_nothing, str_permutations # pylint: disable=no-name-in-module
from modules import shared, shared_items, sd_samplers, ipadapter, sd_models, sd_vae, sd_unet
@@ -127,7 +127,7 @@ axis_options = [
AxisOption("[Refine] Refiner steps", float, apply_field("refiner_steps")),
AxisOption("[Postprocess] Upscaler", str, apply_upscaler, cost=0.4, choices=lambda: [x.name for x in shared.sd_upscalers][1:]),
AxisOption("[Postprocess] Context", str, apply_context, choices=lambda: ["Add with forward", "Remove with forward", "Add with backward", "Remove with backward"]),
- AxisOption("[Postprocess] Face restore", str, apply_face_restore, fmt=format_value),
+ AxisOption("[Postprocess] Detailer", str, apply_detailer, fmt=format_value),
AxisOption("[HDR] Mode", int, apply_field("hdr_mode")),
AxisOption("[HDR] Brightness", float, apply_field("hdr_brightness")),
AxisOption("[HDR] Color", float, apply_field("hdr_color")),
diff --git a/scripts/xyz_grid_shared.py b/scripts/xyz_grid_shared.py
index 586070658..ee8977f52 100644
--- a/scripts/xyz_grid_shared.py
+++ b/scripts/xyz_grid_shared.py
@@ -237,17 +237,17 @@ def apply_context(p: processing.StableDiffusionProcessingTxt2Img, opt, x):
shared.log.debug(f'XYZ grid apply resize-context: "{x}"')
-def apply_face_restore(p, opt, x):
+def apply_detailer(p, opt, x):
opt = opt.lower()
if opt == 'codeformer':
is_active = True
- p.face_restoration_model = 'CodeFormer'
+ p.detailer_model = 'CodeFormer'
elif opt == 'gfpgan':
is_active = True
- p.face_restoration_model = 'GFPGAN'
+ p.detailer_model = 'GFPGAN'
else:
is_active = opt in ('true', 'yes', 'y', '1')
- p.restore_faces = is_active
+ p.detailer = is_active
shared.log.debug(f'XYZ grid apply face-restore: "{x}"')