magliente / computer-vision-demo

Computer-Vision-Demo

The jupyter notebook (.ipynb) serves as a demonstration of previous Computer Vision work I have contributed towards. This notebook is adapted from a final group project for University of California, Berkeley's Computer Vision class.

This project analyzes Imagenette images to investigate the synergy between image pre-processing, feature extraction, and classifiers in achieving accurate and timely image classifications. This research contributes to the growing field of Computer Vision but also captures vast real-world applications with its aim at object identification within images. We explore feature extraction methods and leverage Exploratory Data Analysis techniques, Data Visualization and Data Analysis to present results.

Our best results used Resnet, where we had almost perfect predictions on testing data.