Fogsight is an AI-powered animation engine that transforms abstract concepts into high-quality, narrative animations with bilingual narration and cinematic visuals. It is designed for users who want to easily generate and iterate on animations through a conversational interface, making AI animation accessible to a broader audience.

How It Works

Fogsight leverages Large Language Models (LLMs) for intelligent orchestration of the entire animation pipeline. The core approach involves an LLM interpreting user prompts to generate narration, select visual elements, and define dynamic effects, creating a cohesive animation from a simple text input. This LLM-driven orchestration automates complex creative processes, enabling rapid, high-fidelity animation generation.

Quick Start & Requirements

• Installation: Clone the repository, install dependencies via pip install -r requirements.txt, configure API keys by copying and editing demo-credentials.json to credentials.json, and run with python start_fogsight.py.
• Prerequisites: Python 3.9+, a modern web browser, and an LLM API key (Google Gemini 2.5 Pro recommended).
• Docker: Available via docker-compose up -d after cloning and configuring credentials.json.
• Setup: Requires LLM API key configuration.
• Links: English | 中文

Highlighted Details

• Generates full narrative animations with bilingual narration and cinematic quality from text prompts.
• Features a Language User Interface (LUI) for iterative refinement through multi-turn conversations with the AI.
• Supports OpenAI-compatible SDKs and OpenRouter, with a strong recommendation for Google Gemini 2.5 Pro.
• Demonstrates capabilities with examples like "The Law of Increasing Entropy" and "Bubble Sort."

Maintenance & Community

The project is a member of the WaytoAGI open-source initiative. Contributors include researchers from institutions like the Chinese University of Hong Kong (Shenzhen) and Nanyang Technological University, as well as community members. Contact and community group links are provided.

Licensing & Compatibility

Licensed under the MIT License. This permissive license allows for commercial use and integration into closed-source projects, with a request for attribution.

Limitations & Caveats

The project relies on external LLM API keys, which may incur costs. While Gemini 2.5 Pro is recommended, compatibility with other models via OpenRouter is supported, though performance may vary.