comfyui-outfit

ComfyUI Outfit Selection

A comprehensive outfit generation system for ComfyUI with AI-powered prompt enhancement and dynamic outfit composition.

📖 Table of Contents

• Features
• Prerequisites
• Installation
• Quick Start
• Data & Extensibility
• Project Structure
• License

🎯 Features

👗 Dynamic Outfit Generation

• Female & Male Outfit Nodes: Generate complete outfits with randomization
• Extensive Customization: 20+ outfit categories (torso, legs, feet, accessories, etc.)
• Smart Randomization: Intelligent random selection with seed control
• JSON-Based Data: Easily customizable outfit options via JSON files

🤖 AI-Powered Prompt Enhancement

• Ollama LLM Integration: Enhance prompts with local AI models
• Vision Node Support: Describe images using vision-language models
• Flexible Styling: Multiple prompt styles and creative modes
• Professional Quality: Production-ready prompt generation

🎨 Advanced Customization

• Seed Management: Fixed, random, increment, and decrement modes
• Style Controls: Art styles, lighting, cameras, and more
• Background Options: Diverse scene and environment settings
• Age & Demographics: Comprehensive character customization
• Smart Presets: Curated presets per gender that auto-fill only empty fields
• Makeup UI: Animated, toggleable makeup editor with duplicate/clear controls
• Avoid Terms: Add negative keywords (e.g., blurry, low-res) to refine outputs
• Preset Export: Toggle to print a JSON snippet of your current choices to console

📦 Installation

Method 1: Git Clone (Recommended)

cd ComfyUI/custom_nodes/
git clone https://github.com/manifestations/comfyui-outfit.git

Method 2: Manual Download

1. Download the latest release from the Releases page
2. Extract to ComfyUI/custom_nodes/comfyui-outfit/

Method 3: ComfyUI Manager

1. Open ComfyUI Manager
2. Search for "ComfyUI Outfit Selection"
3. Click Install

🚀 Quick Start

Basic Usage

1. Start ComfyUI
2. Look for nodes under 👗 Outfit category:
   • Female Outfit Node: Generate female outfits
   • Male Outfit Node: Generate male outfits
   • Ollama Prompter: Enhance prompts with AI
   • Simple Ollama: Basic prompt enhancement

Example Workflow

[Female Outfit Node] → [Text Output] → [Your Image Generation Model]

Using Presets

• Choose a Preset from the dropdown. It fills only fields that are empty/none/random.
• You can still override any field afterwards.
• Presets include pose, background, and scene highlights (mood, time, color scheme), plus optional makeup.

Notes:

• If a preset references an unknown key, the node logs a warning to help curate data.
• Presets are loaded from data/presets.json under keys by gender.

Makeup Editor

Avoid Terms

• Use the Avoid field to list items you don’t want (comma-separated or free text).
• These are appended as “Avoid: ...” in the final prompt.

Export Preset

• Toggle “export_preset” to print a preset JSON snippet to the console.
• Paste under your gender in data/presets.json and give it a name.
• Click the Makeup field to open the editor.
• Add items like lipstick, eyeliner, blush, highlighter, etc.
• Use the Enable/Duplicate/Clear All controls for quick edits.

📋 Prerequisites

Core Requirements

• ComfyUI (latest version recommended)
• Python 3.8+

Optional Dependencies

For AI prompt enhancement features, these packages are required. They can be installed by running:

pip install -r requirements.txt

For vision node support:

• An Ollama installation is required.
• You will also need to download vision models (e.g., llava, moondream).

📦 Installation

MIT License. See LICENSE.

🧩 Data & Extensibility

Add or Edit Outfit Options

• Edit JSON files in data/outfit/<gender>/*.json. Each file contains an attire array of objects with a type string.
• The node automatically discovers parts and prepends control values: none, random.

Add Presets

• Edit data/presets.json:
  • Top-level keys: female, male.
  • Under each, add presets by name mapping to { part_or_option_key: value }.
  • Only keys present as inputs will be applied; others are ignored with a warning.
• Keep values aligned with available options in the corresponding JSONs (poses/backgrounds included).

Scene/Style Controls

• Tunables include: mood, time_of_day, weather, color_scheme, description_style, and creative_scale.
• Styles are defined under data/styles/ and are easy to extend.

License

This project is licensed under the MIT License. See the LICENSE file for details.

📁 Project Structure

The project is organized as follows:

• nodes/: Contains the core ComfyUI node implementations.
• utils/: Shared utilities for data loading and prompt building.
• data/: JSON files for outfits, presets, and styles.
• web/: Custom UI components (CSS, JS).
• tests/: Unit tests for the project.
• __init__.py: Registers the custom nodes with ComfyUI.
• requirements.txt: Lists the Python dependencies.