Latent Diffusion is a framework for high-resolution image synthesis that performs the denoising process within a compressed latent space. It uses variational autoencoders to encode images into a lower-dimensional representation, reducing the computational cost of noise prediction compared to operating on raw pixels. The project enables text-to-image generation by integrating natural language descriptions through cross-attention conditioning. It also supports image inpainting and restoration, filling masked or missing image areas with generated content, and example-based synthesis using retrie
SwinIR is a deep learning image restoration framework that uses Swin Transformer architectures to recover image quality. It is designed to restore degraded images by removing noise, blur, and compression artifacts while increasing pixel density. The model provides specialized capabilities for image super-resolution, image denoising, and image deblurring. It also includes a dedicated tool for the removal of JPEG compression artifacts to restore visual quality lost during encoding. The system focuses on improving overall visual fidelity through resolution upscaling, noise removal, and the reco
CVPR 2022 StyleSwin: Transformer-based GAN for High-resolution Image Generation
CVPR 2022 Styleformer - Official Pytorch Implementation
PyTorch Implementation of "Light Field Image Super-Resolution with Transformers"
The main features of zhengyuliang24/lft are: Image Restoration and Generation.
Open-source alternatives to zhengyuliang24/lft include: compvis/latent-diffusion — Latent Diffusion is a framework for high-resolution image synthesis that performs the denoising process within a… jeeseung-park/styleformer — [CVPR 2022] Styleformer - Official Pytorch Implementation. jingyunliang/swinir — SwinIR is a deep learning image restoration framework that uses Swin Transformer architectures to recover image… microsoft/styleswin — [CVPR 2022] StyleSwin: Transformer-based GAN for High-resolution Image Generation. raywzy/ict — High-Fidelity Pluralistic Image Completion with Transformers (ICCV 2021). researchmm/aot-gan-for-inpainting — [TVCG'2023] AOT-GAN for High-Resolution Image Inpainting (codebase for image inpainting).