Image De-Raining using Deconvolutional Networks

Dec 1, 2017
PDF Code

This project was a part of the Neural Networks and Fuzzy Logic course offered at BITS Pilani, K K Birla Goa Campus. The goal of this project was to train a neural network to take as input rainy images and produce rain-free images as output.

As part of this project:

  • Different models (auto-encoders/GANs) were trained to produce rain free images in a supervised setting
  • TensorFlow framework used to implement deep networks for model training
  • Considerable rain-free images were produced both on synthetic and natural rainy images without much perturbations
Deep Learning
Shivin Thukral
Shivin Thukral
Machine Learning Engineer

Working as an MLE on building recommendation systems using ML and NLP techniques

Related