Skip to content

Latest commit

 

History

History
217 lines (172 loc) · 5.86 KB

README.md

File metadata and controls

217 lines (172 loc) · 5.86 KB

Hitchcock - AI-Powered Movie Production Pipeline

Hitchcock is an innovative multi-agent system that automates the movie production pipeline from script to screen. Built with modern AI technologies, it orchestrates specialized agents to handle different aspects of movie production, from scriptwriting to visual implementation.

🎯 Key Features

  • Automated Script Generation: AI-powered script creation with deep research capabilities
  • Intelligent Story Boarding: Automated scene breakdown and shot planning
  • Visual Element Planning: Comprehensive planning of lighting, props, and atmosphere
  • Multi-Agent Architecture: Specialized agents working together in a coordinated pipeline
  • Modern Tech Stack: Built with Python 3.8+, using cutting-edge AI and media processing libraries

🏗 System Architecture

Hitchcock Architecture

The system consists of the following specialized agents:

1. Script Writer Agent

  • Purpose: Creates and analyzes movie scripts
  • Capabilities:
    • Deep web research for historical/cultural context
    • Similar movie analysis for inspiration
    • Scene and dialogue generation
    • Script structure analysis
  • Tools:
    • Web browsing and research tools
    • Text analysis and inspection
    • Scene generation and analysis

2. Story Boarding Agent

  • Purpose: Converts scripts into detailed visual plans
  • Capabilities:
    • Scene importance analysis
    • Shot sequence planning
    • Visual element specification
    • Camera angle and movement planning
  • Features:
    • Automatic scene breakdown
    • Shot-by-shot planning
    • Lighting and atmosphere specification
    • Prop and special effect planning

3. Director of Photography (DOP) Agent

  • Purpose: Handles visual implementation of story boards
  • Status: In development
  • Planned Features:
    • Text-to-image generation
    • Scene composition
    • Visual continuity management
    • Lighting implementation

🚀 Getting Started

Prerequisites

  • Python 3.8 or higher
  • uv package manager

Installation

  1. Install uv:
pip install uv
  1. Clone the repository:
git clone https://github.com/yourusername/hitchcock.git
cd hitchcock
  1. Set up the environment:
uv venv
source .venv/bin/activate  # On Unix
# or
.venv\Scripts\activate  # On Windows
  1. Install dependencies:
uv pip sync requirements.txt

Running the System

  1. Configure environment variables:
cp .env.template .env
# Edit .env with your API keys and configurations
  1. Start the control plane:
python control_plane.py

📁 Project Structure

hitchcock/
├── control_plane.py      # Main orchestration system
├── requirements.txt      # Project dependencies
├── .env.template        # Environment variable template
├── agents/
│   ├── script_writer/   # Script Writer Agent
│   │   ├── agent.py     # Agent implementation
│   │   ├── prompt.py    # Agent prompts
│   │   ├── tools.py     # Script writing tools
│   │   └── scripts/     # Additional scripts
│   ├── story_boarder/   # Story Boarding Agent
│   │   ├── prompt.py    # Agent prompts
│   │   ├── storage.py   # State management
│   │   └── tools.py     # Story boarding tools
│   └── dop/            # DOP Agent (in development)
├── assets/             # Project assets
└── downloads/          # Downloaded research materials

🛠 Development

Managing Dependencies

Add new dependencies:

uv pip install package_name
uv pip compile  # Update requirements.txt

Update all dependencies:

uv pip compile --upgrade
uv pip sync

Test with minimum versions:

uv pip compile --resolution=lowest
uv pip sync

Note: This is an active development project. Features and capabilities may change as development progresses.

Hitchcock: AI-Driven Video Generation Pipeline

Hitchcock is a sophisticated multi-agent movie maker that transforms storyboard descriptions into cinematic videos using state-of-the-art AI models.

Architecture

Core Components

  1. Storyboard Engine

    • Processes structured scene descriptions
    • Manages character continuity and visual themes
    • Uses Pydantic for robust data validation

  2. Image Generation Service

    • Leverages Stable Diffusion XL for high-quality image generation
    • Maintains visual consistency across frames
    • Handles character and environment rendering

  3. Video Processing Pipeline

    • Converts image sequences to video
    • Implements cinematic camera movements
    • Manages temporal consistency

  4. Asset Management

    • Stores and retrieves generated assets
    • Handles metadata and relationships
    • Integrates with VideoDB for efficient storage

Tech Stack

  • Framework: Modal (Serverless deployment)
  • Image Generation: HuggingFace Diffusers
  • Video Processing: MoviePy
  • Data Validation: Pydantic v2
  • Asset Storage: VideoDB

Setup

# Clone the repository
git clone https://github.com/ayushnangia/Hitchcock.git
cd Hitchcock

# Create virtual environment
python -m venv venv
source venv/bin/activate  # On Windows: .\venv\Scripts\activate

# Install dependencies
pip install -r requirements.txt

Project Structure

├── hitchcock/
│   ├── core/
│   │   ├── models.py      # Pydantic models
│   │   └── config.py      # Configuration management
│   ├── services/
│   │   ├── image.py       # Image generation service
│   │   ├── video.py       # Video processing service
│   │   └── storage.py     # Asset management
│   └── utils/
│       ├── validators.py  # Custom validators
│       └── helpers.py     # Utility functions
├── tests/
├── requirements.txt
└── README.md

License

MIT