NORMA eResearch @NCI Library

Enhancing Image Reconstruction with Prediction model using Deep Convolutional GANs

Bhujbal, Sahil (2020) Enhancing Image Reconstruction with Prediction model using Deep Convolutional GANs. Masters thesis, Dublin, National College of Ireland.

[thumbnail of Master of Science]
PDF (Master of Science)
Download (1MB) | Preview
[thumbnail of Configuration manual]
PDF (Configuration manual)
Download (1MB) | Preview


Image Inpainting has gained a lot of popularity within the applications of Computer Vision and Image Processing due to the ability of generating, restoring and modifying images and videos. Current Image Inpainting methods which are incorporated with Generative Adversarial Networks (GANs) can generate feasible inpainting results. However, previous methods are susceptible to fall into the ModeCollapse situation due to inadequate training. This research aims to avoid the situation of ModeCollapse during the training phase by proposing a novel Image Prediction model along with the Deep Convolutional GAN (DCGAN) model. Upon evaluation of the model, results can be backpropagated to the DCGAN model to enhance training, prediction and inpainting of images. Additionally, three datasets are selected for performing experiments based on three different scenarios related to the complexity of images. All images are evaluated in the prediction phase with the help of Inception Score metric and model losses in the training and inpainting phase. Experimental results provide good insights regarding image prediction, thus proving beneficial in enhancing the reconstruction technique.
Index Terms: Image Inpainting, Deep Convolutional Generative Adversarial Network (DCGAN), Image Prediction, Inception Score

Item Type: Thesis (Masters)
Subjects: Q Science > QA Mathematics > Electronic computers. Computer science
T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science
Q Science > QA Mathematics > Computer software
T Technology > T Technology (General) > Information Technology > Computer software
Divisions: School of Computing > Master of Science in Data Analytics
Depositing User: Dan English
Date Deposited: 19 Jan 2021 12:36
Last Modified: 19 Jan 2021 12:36

Actions (login required)

View Item View Item