https://github.com/avhirupc/semantic-image-completion

Implementation of : Semantic Image Inpainting with Perceptual and Contextual Losses Raymond
https://github.com/avhirupc/semantic-image-completion

computer-vision deep-learning deep-neural-networks image-inpainting machine-learning python tensorflow-experiments

Last synced: 4 months ago
JSON representation

Implementation of : Semantic Image Inpainting with Perceptual and Contextual Losses Raymond

• Host: GitHub
• URL: https://github.com/avhirupc/semantic-image-completion
• Owner: avhirupc
• Created: 2017-02-07T12:39:48.000Z (almost 8 years ago)
• Default Branch: master
• Last Pushed: 2020-03-04T09:43:52.000Z (almost 5 years ago)
• Last Synced: 2024-10-10T08:42:29.061Z (4 months ago)
• Topics: computer-vision, deep-learning, deep-neural-networks, image-inpainting, machine-learning, python, tensorflow-experiments
• Language: Jupyter Notebook
• Size: 366 KB
• Stars: 23
• Watchers: 2
• Forks: 9
• Open Issues: 2
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# Semantic Inpainting using DCGANs

* * *

This is an experimental tensorflow implementation of semantic inpainting from corrupted images using DCGANs from the paper [Semantic Image Inpainting with Perceptual and Contextual Losses](https://arxiv.org/abs/1607.07539). A major help was Brandon Amos blog on [Image Completion](https://bamos.github.io/2016/08/09/deep-completion/).One of the major difference between is the training method used.I have used Adam Optimizer instead of gradient descent.

* * * *

## Requirements

* Tensorflow

* glob

* Python 3

* * * *

## Dataset

* I have used Celebrity faces dataset [CelebA](http://mmlab.ie.cuhk.edu.hk/projects/CelebA.html).Download the aligned version ,extract in the same directory as the code

* * * *

## Model Architecture

![alt-text](images/model.png)

* * *

## Few Results are Partial Training

>Note: Due to unavailabity of GPU,i didnt train the model for long.This results are after an hour of training

![alt-text](images/1.jpg) ![alt-text](images/63.jpg)