JIA-Lab-research/outpainting_srn
Wide-Context Semantic Image Extrapolation, CVPR2019
Employs a small-to-large progressive generation scheme with context normalization and relative spatial variant loss to extrapolate semantically coherent content beyond image boundaries. Built on TensorFlow 1.6+ with pretrained models for faces, bodies, and scenes (CelebA-HQ, Cityscapes, Paris streetview). Training uses two-stage optimization: initial reconstruction pretraining followed by adversarial fine-tuning with identity-MRF loss for photorealistic consistency.
135 stars. No commits in the last 6 months.
Stars
135
Forks
32
Language
Python
License
MIT
Category
Last pushed
Sep 03, 2022
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/JIA-Lab-research/outpainting_srn"
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