SpatialAttentionGAN - GitLite

KO | EN

gitlite — search

#react #ai #android #machine-learning #rust #linux #nodejs #macos #java #windows #awesome #kubernetes #mcp #ios

SpatialAttentionGAN

(★ 104)

SaGAN PyTorch "Generative Adversarial Network with Spatial Attention for Face Attribute Editing"

File Explorer

Download Latest Version (.zip)

- 4_attr_results.jpg
data.py
generate.py
LICENSE
README.md
requirements.txt
sagan.py
train.py

# Installation Guide

1. Get the code

git clone https://github.com/elvisyjlin/SpatialAttentionGAN

Downloads the entire project code from GitHub to your computer.

cd SpatialAttentionGAN

Moves into the project folder you just downloaded.

2. Python

Easy Recommended

Prerequisites

Git Needed to download the project code from GitHub.
Python 3 On Windows, be sure to check 'Add Python to PATH' during installation.

pip3 install -r requirements.txt

Installs the Python libraries listed in requirements.txt (or similar).

✅ If it runs without errors and prints output in the terminal, it worked.

Pulled directly from this repo's README.

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# Spatial Attention Generative Adversarial Network

This repository contains the PyTorch implementation of the ECCV 2018 paper "Generative Adversarial Network with Spatial Attention for Face Attribute Editing" ([pdf](http://openaccess.thecvf.com/content_ECCV_2018/papers/Gang_Zhang_Generative_Adversarial_Network_ECCV_2018_paper.pdf)).

My results with images and attention masks on CelebA 128 _(original, eyeglasses, mouth_slightly_open, no_beard, smiling)_

![Results](https://github.com/elvisyjlin/SpatialAttentionGAN/blob/master/pics/4_attr_results.jpg)

## Requirements

* Python 3.5
* PyTorch 1.0.0

```bash
pip3 install -r requirements.txt
```

The training procedure described in paper takes 5.5GB memory on a single GPU.

* Datasets
  * [CelebA](http://mmlab.ie.cuhk.edu.hk/projects/CelebA.html)
    * Put _Align&Cropped Images_ in `./data/celeba/*.jpg`
    * Put _Attributes Annotations_ in `./data/list_attr_celeba.txt`

* Pretrained models (download from http://bit.ly/sagan-results and decompress the zips to `./results`)
  ```
    results
    ├── celeba_128_eyeglasses
    ├── celeba_128_mouth_slightly_open
    ├── celeba_128_no_beard
    └── celeba_128_smiling
  ```

## Usage

Train a model with a target attribute

```bash
python3 train.py --experiment-name celeba_128_eyeglasses --target-attr Eyeglasses --gpu
```

Generate images from trained models

```bash
python3 generate.py --experiment-name celeba_128_eyeglasses --gpu
```