Regal-Project

Introduction

The Regal-Project aims to develop an application that generates long-form videos from a given topic. The videos will have durations ranging from 2 to 5 minutes and will be available in various aspect ratios, including landscape, square, and 9:6. The application will support video generation in all languages.

System Overview

The Regal-Project is designed to provide users with a seamless experience in generating videos from topics. The system will leverage advanced AI models to ensure high-quality video content that meets user specifications in terms of duration, aspect ratio, and language.

Functional Requirements

• As a User, I should be able to input a topic to generate a video.
• As a User, I should be able to specify the duration of the video between 2 to 5 minutes.
• As a User, I should be able to choose the aspect ratio of the video (landscape, square, 9:6).
• As a User, I should be able to generate videos in any language.
• As a User, I should be able to preview the generated video before finalizing.
• As a User, I should be able to download the generated video in a preferred format.

User Personas

• General User: Individuals looking to create videos from topics for personal or professional use. They require an easy-to-use interface and support for multiple languages and formats.

Core User Flows

1. User inputs topic -> selects video duration -> chooses aspect ratio -> selects language -> previews video -> downloads video.
2. User inputs topic -> selects video duration -> previews video -> modifies settings if needed -> downloads video.

Visuals Colors and Theme

• primary: #1A73E8 (Deep Blue)
• primary_light: #E8F0FE (Light Blue)
• secondary: #FF6F61 (Coral)
• accent: #FFD700 (Gold)
• highlight: #FFA500 (Orange)
• bg: #FFFFFF (White)
• surface: rgba(250, 250, 250, 0.8)
• text: #333333 (Dark Gray)
• text_muted: #777777 (Light Gray)
• border: rgba(200, 200, 200, 0.5)

Signature Design Concept

The homepage of the Regal-Project will feature an interactive video reel concept. As users scroll, they will see a dynamic filmstrip that showcases different video styles and formats. Each frame on the filmstrip can be clicked to expand into a full preview of a sample video. The filmstrip will animate smoothly using framer-motion for spring physics, giving users a tactile sense of interacting with a real film reel. This interactive element will provide an engaging way for users to explore the capabilities of the video generator.

Interaction Model & Motion Direction

The landing page will use a "parallax" interaction model. Layers of video frames and UI elements will move at different speeds to create a sense of depth as users scroll. This will be complemented by animated transitions and hover effects to enhance user engagement. Internal pages will adopt a "static" model to prioritize content clarity and ease of use.

Non-Functional Requirements

• The application must support high concurrency to handle multiple video generation requests simultaneously.
• The system should ensure data privacy and secure access control for user-generated content.
• The application should be responsive and perform optimally across various devices and screen sizes.

Tech Stack

• Frontend: React for Web
• Backend: Python, FastAPI
• Database: MongoDB
• AI Models: Google Nano banana for image generation
• AI Tools: Langchain
• Local Orchestration: Docker, docker-compose
• Server-side Orchestration: Kubernetes

Assumptions and Constraints

• The application assumes users have a stable internet connection for video generation and download.
• The system is constrained by the processing power available for real-time video rendering and AI model execution.

Glossary

• Aspect Ratio: The proportional relationship between the width and height of a video.
• Parallax: A design technique where background images move slower than foreground images, creating an illusion of depth.
• Framer-motion: A library for creating animations in React applications.