Back to Space
// Computer Vision / ML

Contestant Disposer

Sports celebrity image classification system using OpenCV and Machine Learning.

PythonOpenCVScikit-learnFlaskDropzone.js

System Architecture Flow

Image Upload

User uploads a photo via Dropzone.js frontend.

Preprocessing

OpenCV detects face/eyes and applies Wavelet Transform.

Classification

SVM model predicts the celebrity identity from extracted features.

Result Display

API returns the result with confidence scores to the UI.

What we have used

  • OpenCV for face and eye detection to crop and preprocess images.

  • Wavelet Transform for feature extraction from images.

  • SVM (Support Vector Machine) classifier for identifying specific athletes.

  • Flask backend to process image uploads and return classification results.

  • Dropzone.js for a smooth drag-and-drop image upload experience.

Uses & Applications

  • Automatically identifies famous sports personalities (e.g., Messi, Federer, Serena Williams) from photos.

  • Can be integrated into sports news portals or fan engagement platforms.

  • Showcases image preprocessing and classification techniques in a web environment.

Future Roadmap

  • Expanding the dataset to include a wider range of athletes and public figures.

  • Implementing Deep Learning (CNNs) to handle lower-quality or partially obscured images.

  • Developing a mobile-responsive interface for on-the-go celebrity identification.