Ha0Tang/LGGAN
[CVPR 2020] Local Class-Specific and Global Image-Level Generative Adversarial Networks for Semantic-Guided Scene Generation
This project helps generate realistic images from basic semantic layouts or translate images from one perspective to another. You provide a segmentation map (like a colored drawing where each color represents an object type, e.g., sky, road, building) and it creates a detailed, high-resolution image, or you give it a bird's-eye view and it generates a street-level view. It's useful for researchers or artists in fields like computer vision, urban planning, or virtual reality who need to synthesize scenes or translate image perspectives.
144 stars. No commits in the last 6 months.
Use this if you need to create realistic outdoor scenes or translate between different perspectives (like aerial to ground views) from simple semantic guides or existing images.
Not ideal if you're looking for a tool to generate abstract art, modify human faces, or perform object recognition, as its focus is on scene synthesis and view translation.
Stars
144
Forks
13
Language
Python
License
—
Category
Last pushed
Feb 18, 2023
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/Ha0Tang/LGGAN"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
yunjey/domain-transfer-network
TensorFlow Implementation of Unsupervised Cross-Domain Image Generation
taesungp/contrastive-unpaired-translation
Contrastive unpaired image-to-image translation, faster and lighter training than cyclegan (ECCV...
PaddlePaddle/PaddleGAN
PaddlePaddle GAN library, including lots of interesting applications like First-Order motion...
tohinz/ConSinGAN
PyTorch implementation of "Improved Techniques for Training Single-Image GANs" (WACV-21)
sagiebenaim/DistanceGAN
Pytorch implementation of "One-Sided Unsupervised Domain Mapping" NIPS 2017