ehsanik/SeGAN
SeGAN: Segmenting and Generating the Invisible (https://arxiv.org/pdf/1703.10239.pdf)
This project helps computer vision researchers and developers predict what is hidden behind objects in an image. By analyzing visible parts of a scene, it can generate the appearance of occluded areas and segment both visible and invisible parts of objects. The input is an image with occluded objects, and the output is a completed image with predicted invisible regions and their segmentations.
No commits in the last 6 months.
Use this if you need to infer the complete appearance and boundaries of objects that are partially hidden in images.
Not ideal if your application requires real-time processing or is not focused on indoor scenes with clear occluder-occludee relationships.
Stars
68
Forks
12
Language
Lua
License
—
Category
Last pushed
Nov 17, 2021
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/ehsanik/SeGAN"
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