🌐 Emotional Sphere Visualization Engine

A high-performance Python application that transforms text into real-time 3D emotional simulations. By leveraging the Google Gemini API for semantic analysis and ModernGL for GPU-accelerated rendering, the system generates a morphing sphere that represents human feelings through complex vertex displacement and lighting.

1.mp4

8.mp4

🚀 Overview

This project bridges the gap between Natural Language Processing (NLP) and Computer Graphics. It breaks down user input into emotional fragments based on Plutchik's Wheel of Emotions, mapping them to physical parameters like vibration, roughness, and color in a 3D environment.

🛠 Tech Stack

Category	Tools
Language	Python 3.13.5, GLSL (OpenGL Shading Language)
Graphics	ModernGL, GLFW, PyRR (Matrices), NumPy
AI / NLP	Google Generative AI (Gemini Pro), python-dotenv
UI / UX	CustomTkinter
Vision/Output	OpenCV (MP4 Recording support)

🏗 Architecture

The application follows a Two-Stage Interaction Pipeline to ensure stability between the event-driven UI and the high-frequency rendering loop:

1. Input Stage (UI): Built with CustomTkinter, users input text and configure recording options.
2. Analysis Stage (Client): The text is sent to the Gemini API with a strict system prompt. The AI returns a structured JSON-like response identifying primary/secondary emotions and their intensities.
3. Simulation Stage (GPU): The system transitions to a GLFW window where ModernGL handles the rendering.
4. Interpolation Engine: Emotions are not static; the system uses weighted linear interpolation to create smooth transitions between complex emotional states (e.g., blending "Joy" with "Anticipation").

🎨 Emotional Mapping Logic

The sphere's behavior is dictated by three main physical variables controlled via Vertex Shaders:

• Speed: Increases the frequency of the deformation wave (higher arousal).
• Roughness: Adds complexity and "peaks" to the surface (Perlin Noise frequency).
• Distortion: Defines the amplitude/size of the displacement.

Emotional Color Palette & Dynamics

The system follows a logic where intensity equals saturation.

• Joy (Yellow): Calm vertical movements.
• Rage (Red): High-frequency "spike" explosions via spherical Perlin noise.
• Sadness (Blue): Slow, descending wave patterns mimicking tears.
• Fear (Green): Shaking, constricted "clamped" movements.

⚡ Key Features

• Complex Shaders: Uses Displacement Mapping and Phong Reflection Models for realistic light interaction.
• Hybrid Emotions: Detects "Nuances" (Matices). For example, Joy + Acceptance = Love; Anger + Alertness = Aggressiveness.
• Post-Processing: Real-time Gaussian blur filters applied via Framebuffer Objects (FBO).
• Recording: Integrated OpenCV functionality to export the simulation as an MP4 file.

🔧 Requirements

Only theese libraries below are required to run the engine + Google Cloud API key compatible with the Gemini API

pip install moderngl glfw numpy pyrr opencv-python customtkinter google-generativeai python-dotenv