Create Offline AI Tool (EXE) for Orthodox Wood-Carved Icon Style Batch Processing

Description:
I need a Windows offline tool (EXE) that can batch-process a folder of my images into an Orthodox icon style that looks like a wood-carved engraving made from cherry wood with a natural varnish highlighting the grain and texture.

Style Prompt:
"Orthodox icon in the style of wood-carved engraving, carved in cherry wood with natural varnish that highlights the texture and grain of the wood. In the center, prominently displayed, is the image provided, with clear modern Greek letters. The frame is simple and plain, without intricate carvings, to emphasize the figures of the Saints."

Requirements:

Completely offline (no API calls) using Stable Diffusion + Automatic1111 WebUI or similar

Ability to select an input folder and process all images inside automatically

Apply the same style prompt to every image for consistent results

Output high-resolution PNG or JPEG images in an output folder

Include all models, LoRAs, and configurations inside the EXE package so it works without installation

Simple interface (button: "Select Folder" → "Process")

Deliverables:

Windows EXE that works offline

All required AI models included locally

Full source code (Python or other language)

Skills Required:

Stable Diffusion setup (Automatic1111 or ComfyUI)

LoRA/custom model usage for wood-carved iconography

Python scripting & batch automation

EXE packaging (PyInstaller or similar)

Knowledge of Greek typography (preferred)

2️⃣ Python Κώδικας για τον Freelancer (Batch Processing + Offline)
python
Αντιγραφή
Επεξεργασία
import os
import base64
import requests

API_URL = "http://127.0.0.1:7860/sdapi/v1/img2img"

PROMPT = (
"Orthodox icon in the style of wood-carved engraving, carved in cherry wood "
"with natural varnish that highlights the texture and grain of the wood. "
"In the center, prominently displayed, is the image provided, with clear modern Greek letters. "
"The frame is simple and plain, without intricate carvings, to emphasize the figures of the Saints."
)

NEGATIVE_PROMPT = "blurry, distorted, low quality, extra limbs, bad hands"

INPUT_FOLDER = "input_images"
OUTPUT_FOLDER = "output_images"
os.makedirs(OUTPUT_FOLDER, exist_ok=True)

for filename in os.listdir(INPUT_FOLDER):
if filename.lower().endswith((".png", ".jpg", ".jpeg")):
image_path = os.path.join(INPUT_FOLDER, filename)

with open(image_path, "rb") as f:
image_bytes = f.read()

payload = {
"init_images": [base64.b64encode(image_bytes).decode()],
"prompt": PROMPT,
"negative_prompt": NEGATIVE_PROMPT,
"width": 1024,
"height": 1024,
"steps": 30,
"cfg_scale": 7,
"denoising_strength": 0.5
}

response = requests.post(API_URL, json=payload)
result = response.json()

for i, img_data in enumerate(result["images"]):
out_path = os.path.join(OUTPUT_FOLDER, f"{os.path.splitext(filename)[0]}_carved.png")
with open(out_path, "wb") as f:
f.write(base64.b64decode(img_data))

print("✅ All images processed!")