Chirikku Kutta : A Smile Volume Controller with Peace Sign Challenge 🎯

Author: Mohamed Nihal T K N Team / Single-member Project: Mohamed Nihal T K N

---

Project Summary

A webcam-based desktop utility that dynamically controls system volume based on how wide you smile, and periodically challenges you to show a peace sign and open your mouth wide to verify engagement. If you look away for a set time the app launches a random "distraction" site to get your attention back.

This is a single‑member final project implementing real‑time face and hand tracking using MediaPipe, computer vision with OpenCV, and system audio control (Windows) via pycaw.

---

Why this project?

The Problem (tongue-in-cheek)

People get distracted while using their computers, or their playlists are either too quiet or too loud. Also, existing volume controls are boring.

The Solution

Automatically set the system volume in real time based on a "smile score" (mouth width × mouth opening). To keep the user honest and engaged, the app occasionally mutes itself and demands a peace sign + open mouth challenge — complete it to save a snapshot and restore volume.

---

Features

• Real-time face mesh detection (MediaPipe) to compute a smile score.
• Smooth mapping from smile score → system master volume (Windows via pycaw).
• Hand tracking to detect a peace sign (index + middle fingers extended).
• Random time-based challenge that mutes volume until user performs the gesture and opens mouth wide enough.
• Attention detection — if you look away for a configurable timeout, the app opens a random distraction website.
• Saves snapshot on successful challenge completion.

---

Technologies / Libraries

Software

• Python 3.10.11 (recommended)
• OpenCV (cv2)
• MediaPipe (mediapipe) — Face Mesh & Hands
• NumPy (numpy)
• pycaw (pycaw) — Windows audio control
• comtypes
• ctypes
• Other standard libs: math, time, random, datetime, webbrowser

Platform

• Developed and tested on Windows (pycaw used for system volume). The app will run on other OSes but system volume control will be disabled unless replaced with an OS-specific module.

---

Files

• Chirikku_Kutta.py — Main application (the code you provided).
• README.md — This file.
• snapshot_YYYYMMDD_HHMMSS.png — Generated on successful challenge (saved beside the script).

---

Installation

1.Install Python 3.10.11. If you don't have this specific version, it's highly recommended to install it from the official Python website to ensure compatibility.

2. Clone the repository or copy the files to your machine.

3. (Optional) Create and activate a virtual environment:

python -m venv venv
# Windows
venv\Scripts\activate
# macOS / Linux
source venv/bin/activate

4. Install required packages:

pip install opencv-python mediapipe numpy comtypes pycaw

Note: Installing pycaw and using it requires Windows. If you're on macOS / Linux, the script will still run but volume control will be disabled and a message will be printed.

---

Configuration (in main.py)

You can tweak the following constants near the top of the script:

• SMILE_SCORE_MIN, SMILE_SCORE_MAX — range for mapping smile score → volume.

• ATTENTION_TIMEOUT — seconds to wait before triggering the attention penalty when face is not detected or off-center.

• DISTRACTION_SITES — list of URLs opened when attention is lost.

• Challenge settings:

  • CHALLENGE_MIN_INTERVAL, CHALLENGE_MAX_INTERVAL — how often (seconds) a challenge may appear.
  • CHALLENGE_MOUTH_THRESHOLD — mouth openness threshold for challenge success.
  • CHALLENGE_TIMEOUT — seconds allowed to complete the challenge.

---

Usage

1. Ensure your webcam is connected and not used by another application.
2. Run the app using Python 3.10.11:

# Use the specific python version to run the script
python3.10 Chirikku_Kutta.py

3. The app window shows:

• A live video feed.
• Current attention status (FOCUSED / LOST).
• A vertical volume bar and numeric percentage (when no challenge is active).
• A CHALLENGE overlay when a challenge is active.

4. Controls:

• Press q to quit.

---

Implementation Notes

• Smile score = horizontal mouth corner distance × vertical lip distance. This produces a value interpolated to 0–100% volume.
• Face centering is estimated using nose tip relative to left/right face edges.
• Peace sign detection uses MediaPipe hand landmarks to check whether index and middle fingertips are above their pip joints while ring & pinky are folded.
• The app mutes volume during a challenge (if OS volume control is available) and restores it upon success.

---

Screenshots

---

Troubleshooting & Tips

• If the webcam fails to open, ensure no other app is using it and that your OS permissions allow camera access.
• If pycaw fails to initialize, you'll see a printed error and the script will continue with volume control disabled.
• Tune SMILE_SCORE_MIN/MAX and CHALLENGE_MOUTH_THRESHOLD for your camera distance and facial proportions.
• Lighting helps: even, front-facing light improves detection robustness.

---

Potential Improvements

• Smooth volume changes using a low-pass filter to avoid jitter.
• Add cross-platform volume control (macOS: osascript or coreaudio, Linux: pulseaudio/pactl).
• Add calibration step for smile and mouth thresholds per user.
• Improve attention detection using eye tracking / gaze estimation.
• Add GUI controls to change config at runtime.

---

Demo

Note: The demo video availabe at https://github.com/nighaaaaal/useless_project_temp/blob/main/demo.mp4. However, in the demonstration video, the audio doesnt change as it is a screenrecording, But in it its clearly visible the audio bar changes, so its working, You can try it yourself.

TO VIEW THAT DOWNLOAD THE RAW FILE.

---

License & Acknowledgements

• This project is shared by Mohamed Nihal T K N.
• Libraries used: MediaPipe, OpenCV, NumPy, pycaw.

---

Made with ❤️ by Mohamed Nihal T K N