Edit Mind

Edit Mind is an open-source, self-hosted video indexing and AI-powered search application that runs entirely on your local machine. It analyzes your video library using advanced AI technologies including object detection, face recognition, emotion analysis, and transcription to create searchable metadata. With Edit Mind, you can search through your entire video collection using natural language queries, finding specific scenes, spoken words, or visual content in seconds.

1. Install Docker Desktop on your computer and ensure it is running properly.
2. Create a project directory and download the docker-compose.yml and environment configuration files from the GitHub repository.
3. Configure your .env file with your video folder path, AI model preferences (Ollama, local model, or Gemini API), and security keys.
4. Run "docker compose up" to start all services including the web app, background jobs, ML service, and databases.
5. Access the web interface at http://localhost:3745 and log in with the default credentials (admin@example.com / admin).
6. Navigate to Settings, add your video folder, and click "Rescan" to start indexing your videos with AI-powered analysis.

Local Video Indexing: Automatically watches for new video files and queues them for AI-powered analysis including transcription and frame analysis.

Face Recognition: Detects and recognizes faces in your videos using DeepFace technology for people search capabilities.

Object Detection: Identifies objects, text, and scenes within video frames using computer vision models.

Natural Language Search: Search through your entire video library using conversational queries powered by ChromaDB vector database.

Transcription Support: Extracts spoken words from videos using OpenAI Whisper for text-based searching.

Multi-Model AI Support: Choose between Ollama, local GGUF models, or Google Gemini API for natural language processing.

Docker Containerization: Fully containerized deployment with docker-compose for easy setup on any machine.

#1 Video Editors and Content Creators: Quickly find specific scenes, B-roll footage, or previously recorded content across large video libraries without manually scrubbing through hours of footage.

#2 Media Archives and Libraries: Organizations can index and search through historical video archives, documentary footage, or educational content using natural language queries.

#3 Security and Surveillance Review: Review and search through security camera recordings by describing events, objects, or people you are looking for.

#4 Personal Video Collections: Home users can organize and search family videos, vacation footage, or personal recordings to find memorable moments instantly.

#5 Research and Documentation: Researchers can analyze video interviews, lectures, or field recordings to find specific topics or statements.

Is Edit Mind free to use? Yes, Edit Mind is completely free and open-source under the MIT license. You can self-host it on your own hardware without any subscription fees.

Does Edit Mind require an internet connection? Edit Mind can run fully offline using local AI models (Ollama or GGUF). However, if you choose to use the Gemini API option, an internet connection is required.

What are the hardware requirements? Edit Mind runs in Docker containers and works on any computer with Docker Desktop installed. For GPU acceleration with NVIDIA cards, use the CUDA-enabled docker-compose file.

What video formats are supported? Edit Mind supports common video formats through OpenCV and FFmpeg, including MP4, AVI, MKV, MOV, and other standard formats.

Can I use my own AI models? Yes, Edit Mind supports multiple AI backends including Ollama with custom models, local GGUF model files, or the Google Gemini API.