Tejpal Virdi
about
Papers
Github
06 Dec 2017
Deep Audio Prior
In this post, we look into the recent paper "Deep Image Prior" and how it can be applied to audio enhancement.
25 Nov 2017
Protecting Privacy with Synthetic Training Data using Dual GANs
We explore a stacked GAN architecture to produce synthetic training data that avoids privacy concerns and allows for sufficient training of deep, data-hungry networks.
22 Nov 2017
Super-resolution GAN for Medical Images
Applying Super-resolution to restore quality in colonscopy video.
21 Nov 2017
Towards an Autonomous Car using Computer Vision
Vehicle detection and Depth Estimation are important steps towards building an autonomous car. We utilize a U-Net segmentation network to segment out vehicles given a 2d image and then train a CNN to learn depth from stereo images.
15 Nov 2017
Neural Style Transfer
In this blog post, we review neural style transfer and how it has advanced in the last few years.
03 Nov 2017
The Generative Adversarial Network Explained
In this post, we explore the architecture of the GAN and its ability to generate images no one has seen before.
29 Oct 2017
A Simple Neural Network
Over the last weekend, we committed ourselves to learning exactly how a neural network (NNs) works. This intro blog post will teach you everything you need to get a straightforward neural net running.
29 Oct 2017
Convolutional Neural Networks Explained
We explore the convolutional neural network: a network that excel at image recognition and classification.
05 Oct 2017
Facial Recognition through Eigenfaces
This blog post demonstrates the process of analyzing a set of images with the goal of identifying an average face, prominent face, eigenface, and recognition system using eigenvectors.