Load Image

DWPose Estimator

Load ControlNet Model

Apply ControlNet (Advanced)

Preview Image

KSampler

VAE Encode

Load Checkpoint

ReActor 🌌 Fast Face Swap

Scale To Megapixels

CLIP Text Encode (Prompt)

Depth Anything V2

DownloadAndLoadDepthAnythingV2Model

My Ollama Vision

✨DisplayText_Zho

The photo appears to be a vintage or retro-filtered photograph, characterized by its faded and grainy texture. The image is centered around a young girl sitting on the ground, seemingly outside in front of what looks like a building with a flat roof. She is dressed in casual attire and is gazing off to the side, her posture relaxed with one arm resting on the ground behind her. The girl's expression is calm and contemplative.

The photo captures an everyday moment with a sense of nostalgia, evoking memories of childhood innocence. The composition of the image is quite simple, focusing on the central subject without any additional elements to distract from the girl's presence. The light source seems to be natural, possibly coming from above, which casts soft shadows and highlights the textures within the photograph.

The framing of the photo is square, placing the girl symmetrically in the frame. There is no additional perspective or forced angle that would draw attention away from her. The image uses the rule of thirds to position the girl slightly off-center, which creates a sense of balance and composition without appearing overly structured.

The unique elements that draw the viewer's attention are primarily the faded quality of the photo, which gives it an old-fashioned feel, and the candidness of the young girl, who seems to be lost in thought or perhaps just enjoying a quiet moment outdoors. The simplicity of the scene, coupled with the timeless quality of the photograph, creates a lasting impression on the viewer.

🔤 CR Text

----

Please create an image generation prompt in English less than 46 words to fit the brief above, operate step by step:

Step 1: If there are no characters in the brief, skip this step. 
List the names and genders of all the characters in the result of Step 1. 

Step 2: If there are no characters in the brief, skip this step. 
Use a short phrase to describe the result of Step 1. This phrase can use plurals or and, do not use any punctuation. Do not use any punctuation in this short phrase. The format is as follows:
only one person:
- 1boy
- 1girl
two persons:
- 2boys
- 2girls
- 1girl and 1boy
three or more persons:
...

Step 3: Keep the key information of the brief above, add more detail and nuance to the prompt. 

Format: three-part bracket 
{[result of Step 2(if there is no explicit subject, remove this part)], [detailed features and actions of each subject from left to right, top to bottom(if there is no explicit subject, remove this part)], [scene, atmosphere, etc.]}

examples：
one person:
- {[1girl], [souryuu asuka langley from neon genesis evangelion, rebuild of evangelion], lance of longinus, cat hat, plugsuit, pilot suit, red bodysuit, sitting, crossed legs, black eye patch, throne, looking down, from bottom, looking at viewer, outdoors, (masterpiece), (best quality), (ultra-detailed), very aesthetic, disheveled hair, perfect composition, moist skin, intricate details]}
two persons:
- {[1boy and 1girl], [Chun-Li from Street Fighter and Spider-Man from Marvel movies], [engaged in a fierce battle, dynamic fighting poses, action movie poster]}
- {[1boy and 1girl], [K-pop male singer (left of center) with short black hair, brown eyes, white color, formal suit with satin lapel and bow tie, halfbody shot, K-pop female singer (center) with long straight black hair, dark brown eyes], [white color, formal silver dress, halfbody shot], both singers holding hands up in peace signs, bright colors, sunny background]}
non-human subject:
- {[1cat], [sitting on a stone floor, looking down with a frowning expression, yellow eyes, white coat with grey markings around its face, slightly open mouth as if meowing or purring], [stone floor as the background, neutral lighting, simple composition, textured background, humorous and endearing tone]}
one person and one non-human subject:
- {[1boy and 1cat], [boy with short black hair and brown eyes, cat sitting on a stone floor, looking down with a frowning expression, yellow eyes, white coat with grey markings around its face, slightly open mouth as if meowing or purring], [stone floor as the background, neutral lighting, simple composition, textured background, humorous and endearing tone]}
no explicit subject:
- {[Mighty water plunges, glistening jewels in sunlight. Surrounding greenery frames nature's spectacle. Sound roars, a symphony of life. Paint the power, beauty, and serenity of this breathtaking scene]}

Step 4: Only output the content in {} of result of Step 3, don't output anything else, such as "Here is the result:", "Here is the image generation prompt in English", "Here is the image generation prompt" and "Note: ...".

✨ConcatText_Zho

My Ollama Generate

drawing,paiting,crayon,sketch,graphite,impressionist,noisy,blurry,soft,deformed,ugly,lowers,bad anatomy,bad hands,text,error,missing fingers,extra digit,fewer digits,cropped,worst quality,low quality,normal quality,jpeg artifacts,signature,watermark,username,(monochrome,greyscale,old photo), text

{[1girl], [casual attire, relaxed posture with arm resting on ground behind her, candid expression of calm contemplation, seated on the ground outdoors in front of a building with a flat roof under natural light, vintage filter giving faded and grainy texture], [no additional elements to distract from subject, off-center composition following rule of thirds, soft shadows and highlights accentuating textures]}

Clean GPU Used

Automatic CFG

VAE Decode

Scratch Mask

Restore Old Photos

Load Restore Old Photos Model

Load Scratch Mask Model

Crop Image (Advanced)

Save Image

Round Image (Pad/Crop) (Advanced)

----

Only output the content in {} of the content above,  don't output anything else.

auto prompt

controlnet

input and output

KSampler and ReActor

Bringing Old Photos Back to Life

model

positive

negative

latent_image

LATENT

MODEL

CLIP

clip

text

CONDITIONING

samples

IMAGE

images

MASK

image

POSE_KEYPOINT

CONTROL_NET

control_net

pixels

input_image

source_image

face_model

face_boost

FACE_MODEL

upscale_model_opt

da_model

da_v2_model

description

show_help

text_1

text_2

prompt

response

anything

scratch_model

mask

bopbtl_models

scratch_mask

crop_left_override

crop_right_override

crop_bottom_override

crop_top_override

nearest_x_override

nearest_y_override

split_x_override

split_y_override

pad_value_override

crop_left

crop_right

crop_bottom

crop_top

git+https://github.com/WASasquatch/img2texture.git@a546609a915caf1dcb84a15bf44b5b639ad924e6

duitpower_87490

This workflow uses:-- Remove scratches from photo- Use controlnet depth/openpose & tile- Use Ollama LLM to describe the old photo- Use Reactor to swap faces from old photo to new photo

Restore Old Photo

OpenArt

Workflows

Active Sessions

Restore Old Photo

Description

Discussion

Loading...