pratyakshajha / project-dog-classification

Second project for Udacity Deep Learning Nanodegree

Project Overview

This repository hosts code used in second project on Convolutional Neural Networks (CNN) project in the Deep Learning Nanodegree. In addition to meeting the project specifications, I also added on a few things.

Given an image of a dog, the algorithm will identify an estimate of the canine’s breed. If supplied an image of a human, it will identify the resembling dog breed.

Getting started

Open the dog_app.ipynb on a GPU enabled Python notebook environment, or Google colab (a link is directly on top in the notebook file). The notebook consists of further technical details. Run the notebook to classify your own dog images.

Further improvements

• Implement the same thing with Tensorflow.
• Webapp using flask
• Overlay a Snapchat-like filter with dog ears on detected human heads.
• Added functionality: A single class with highest probability is shown. Added topk classes, if probability is less than a threshold.
• Experiment with multiple dog and human detectors.