CV2_SoSe_19

Repo for maintaining CV2 coursework and project

Project--Saliency System

Build a saliency prediction system using the 1200 training saliency images. The architecture used is similar to ML-Net.

Exercise 2

Training a simple Neural network for linear regression task.

Exercise 3

Training a simple convolutional neural network on CIFAR-10 dataset.

Exercise 4

Transfer Learning using VGG16(Trained on ImageNet 1000 classes) to train the classifer on CIFAR-10 classes.

Exercise 5

Training the CNN for saliency output(essentially image output). using the ML-Net weighted loss function for efficient learning.

Exercise 6

Add Regularization in the model created in ex5 to avoid overfitting