gans-awesome-applications

Curated list of awesome GAN applications and demonstrations.
Note: General GAN papers targeting simple image generation such as DCGAN, BEGAN etc. are not included in the list.

The landmark papers that I respect.

• Generative Adversarial Networks, [paper], [github]
• Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks, [paper], [github]
• BEGAN: Boundary Equilibrium Generative Adversarial Networks, [paper], [github]

---

Applications using GANs

Font generation

• Learning Chinese Character style with conditional GAN, [blog], [github]

Anime character generation

• Towards the Automatic Anime Characters Creation with Generative Adversarial Networks, [paper]
• A simple PyTorch Implementation of Generative Adversarial Networks, focusing on anime face drawing, [github]

Interactive Image generation

• Generative Visual Manipulation on the Natural Image Manifold, [paper], [github]
• Neural Photo Editing with Introspective Adversarial Networks, [paper], [github]

Text2Image (text to image)

• TAC-GAN – Text Conditioned Auxiliary Classifier Generative Adversarial Network, [paper], [github]
• StackGAN: Text to Photo-realistic Image Synthesis with Stacked Generative Adversarial Networks, [paper], [github]
• Generative Adversarial Text to Image Synthesis, [paper], [github], [github]
• Learning What and Where to Draw, [paper], [github]

3D Obejct generation

• Parametric 3D Exploration with Stacked Adversarial Networks, [github], [Youtube]
• Learning a Probabilistic Latent Space of Object Shapes via 3D Generative-Adversarial Modeling, [paper], [github], [Youtube]

Image Editing

• Invertible Conditional GANs for image editing [paper], [github]
• Age Progression/Regression by Conditional Adversarial Autoencoder, [paper], [github]
• CAN: Creative Adversarial Networks Generating “Art” by Learning About Styles and Deviating from Style Norms, [paper]

Human Pose Estimation

• Pose Guided Person Image Generation, [paper]

Domain-transfer (e.g. style-transfer, pix2pix, sketch2image)

• Image-to-Image Translation with Conditional Adversarial Networks, [paper], [github], [Youtube]
• Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks, [paper], [github], [Youtube]
• Learning to Discover Cross-Domain Relations with Generative Adversarial Networks, [paper], [github]
• Unsupervised Creation of Parameterized Avatars, [paper]
• UNSUPERVISED CROSS-DOMAIN IMAGE GENERATION, [paper]
• Precomputed Real-Time Texture Synthesis with Markovian Generative Adversarial Networks, [paper], [github]
• Pixel-Level Domain Transfer [paper], [github]

Image Inpainting (hole filling)

• Context Encoders: Feature Learning by Inpainting, [paper], [github]
• Semantic Image Inpainting with Perceptual and Contextual Losses, [paper], [github]
• SEMI-SUPERVISED LEARNING WITH CONTEXT-CONDITIONAL GENERATIVE ADVERSARIAL NETWORKS, [paper]
• Generative Face Completion, [paper], [github]

Super-resolution

• Image super-resolution through deep learning, [github]
• Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network, [paper], [github]

Image Blending

• GP-GAN: Towards Realistic High-Resolution Image Blending, [paper], [github]

Visual Saliency Prediction (attention prediction)

• SalGAN: Visual Saliency Prediction with Generative Adversarial Networks [paper], [github]

---

Did not use GAN, but still interesting applications.

Real-time face reconstruction

• Model-based Deep Convolutional Face Autoencoder for Unsupervised Monocular Reconstruction, [paper], [github], [Youtube]

Super-resolution

• Learning to Simplify: Fully Convolutional Networks for Rough Sketch Cleanup, [paper], [site link], [Youtube]

Photorealistic Image geneation (e.g. pix2pix, sketch2image)

• The Sketchy Database: Learning to Retrieve Badly Drawn Bunnies, [paper], [Youtube]
• PatchMatch: A Randomized Correspondence Algorithm for Structural Image Editing, [paper], [github], [Youtube]

Human Pose Estimation

• Knowledge-Guided Deep Fractal Neural Networks for Human Pose Estimation, [paper], [github]

---

GANs tutorials with easy and simple example codes for starters.

• 1D Generative Adversarial Network Demo
• starter from "How to Train a GAN?" at NIPS2016
• [paper], [github], [Youtube]

---

Implementations of various types of GANs collection

• nashory/gans-collections.torch, torch7
• hwalsuklee/tensorflow-generative-model-collections, pytorch
• wiseodd/generative-models, both pytorch and tensorflow

Author

Minchul Shin, @nashory