ViewFlow

Version: 1.0.1

Last updated: 2025-12-04

ViewFlow is a modern, lightweight video sharing platform prototype built with Flask. It features a custom-built video player, a sophisticated recommendation engine, and a sleek, theme-aware user interface.

🚀 Key Features

📊 Analytics & Insights

Empower creators with data-driven insights directly on their profile:

• Dashboard: A dedicated "Analytics" tab for channel owners.
• Visual Charts: Interactive graphs showing Views and New Subscribers over the last 30 days.
• Performance Metrics: Track total views, subscriber count, and video library size.
• Top Content: Identify your best-performing videos at a glance.

🎬 Advanced Video Player

A custom-built, themeable HTML5 player that rivals major platforms:

• "Most Replayed" Heatmap: A YouTube-style graph overlay on the seek bar showing the most popular segments of a video.
• Seek Previews: Hover over the timeline to see a thumbnail preview of that exact moment.
• Resolution Switching: Seamlessly switch between 720p, 480p, and 360p qualities.
• Seamless Playback: Plays uploaded videos seamlessly with a modern HTML5 interface.
• Theatre & Fullscreen Modes: Immersive viewing options.
• Ambient Mode: Dynamic background lighting effects based on video content.

🛠️ Content Management

• Background Uploads: Videos process asynchronously (transcoding, thumbnail generation) so you can keep browsing.
• Edit Details: Update video titles, descriptions (with Markdown support), categories, and tags anytime.
• Visibility Control: Instantly toggle videos between Public and Private with a single click.
• Smart Progress: Real-time upload status indicators on your profile icon.

🧠 Machine Learning Recommendation Engine

ViewFlow includes a powerful content-based filtering system that personalizes the viewing experience:

• Vector-Based Profiling: Builds a dynamic user profile vector based on watch history.
• Smart Context Awareness:
  • Recency Decay: Prioritizes recent interests over older history.
  • Replay Boosting: Detects and boosts content you watch repeatedly.
  • "Current Mood": Heavily weights the last 2 videos to adapt instantly to your current session.
• Personalized Feeds: "For You", "From Your Channels", and "Up Next" suggestions.

👤 Enhanced User Experience

• Rich Profiles: Customize your presence with profile pictures, bio, location, and more.
• Social Interactions: Like, dislike, subscribe, and comment asynchronously without page reloads.
• Dark/Light Mode: A beautiful, persistent theme system that respects your eyes.
• Responsive Design: Fully responsive layout that looks great on desktop and mobile.

📦 Quick Start

Prerequisites

• Python 3.10+
• FFmpeg (for video processing)

Installation

1. Clone and Setup

   # Create a virtual environment
   python3 -m venv .venv
   source .venv/bin/activate
   
   # Install dependencies
   pip install -r requirements.txt

2. Run the Server

   python3 test.py

   The server will start on http://127.0.0.1:5000.

   Note: The application will automatically create a viewflow.db database and an uploads/ directory on the first run.

🏗️ Project Structure

• test.py: The main entry point and development server. Contains models and route logic.
• recommendations.py: The core logic for the ML recommendation engine.
• templates/: Jinja2 templates for the frontend.
• static/: CSS, JavaScript, and assets.
• models.py: SQLAlchemy database models.
• views.py: Route definitions (mirrored in test.py for the dev server).

🤝 Contributing

This project is a prototype for educational and development purposes. Feel free to fork, experiment, and submit pull requests!

---

Developed by Gautham Nair & Deepak Patel