Object Detection and Image Classification of Stanford Dogs Dataset

Object Detection

YOLOv3 model was trained here using the annotations to identify the location of a single class(dog) in an image.

Image Classification

Transfer Learning was used to train the classifier of the InceptionV3 model to identify the breed of the dogs apart from the 119 other dog breeds. Link to Kaggle Kernel

Bringing them together

Main notebook shows how the 2 models can be brought together to process an image and identify the dog breed.