Overview / Usage

This example shows how to use transfer learning to retrain ResNet-50, a pretrained convolutional neural network, to classify a new set of images.In this case the network is used to identify objects from the Caltech-101 dataset. Transfer learning is commonly used in deep learning applications. One can take a pretrained network and use it as a starting point to learn a new task. Fine-tuning a network with transfer learning is usually much faster and easier than training a network with randomly initialized weights from scratch.

Methodology / Approach

Object Detection has been achieved by implementing the following steps in MATLAB

• Loading the data from the dataset
• Loading the pretrained network
• Replace the final layers
• Split the data into testing and training data
• Train the network
• Classify Validation images
• Obtain accuracy

Technologies Used

• Matlab 2018b
• Deep Learning Toolbox
• Resnet-50 architecture
• Caltech-101

Repository

https://github.com/ritwikraha/Object-Detection-using-Transfer-Learning