How this analysis was created: This summary and feature list were written by an AI model that read the project's README and public documentation pages. Each feature links to the documentation it came from; stars, license and language come straight from the GitHub API. The model does not read the source code, and the analysis is refreshed when the project is re-analysed. Learn more on our About page.
Lama is an image restoration framework and deep learning model designed for image inpainting and object removal. It provides the tools necessary to train and evaluate neural networks that fill masked areas and repair corrupted visual data. The system utilizes a Fourier convolution neural network to maintain global image structure and reconstruct periodic patterns. This architecture allows for resolution-independent inference, enabling the processing of high-resolution images without increasing memory or computational requirements. The project includes a synthetic dataset generator that creat
Inpainting using Denoising Diffusion Probabilistic Models
Awesome-GANs is a curated resource list and research repository focused on the development and evaluation of generative adversarial networks. It serves as a structured index for academic literature and open-source implementations dedicated to the creation of synthetic data generators. The project provides a framework for training competing neural networks to produce outputs that mimic the statistical properties of original datasets. It emphasizes the use of configuration-driven pipelines to manage model hyperparameters and dataset paths, facilitating reproducible research workflows and standa
Keras-GAN is a collection of generative adversarial network implementations built with Keras for synthetic data generation and image manipulation. It provides frameworks for image-to-image translation, image inpainting, and neural image super-resolution. The library includes tools for learning disentangled latent space representations to control specific attributes of synthetic outputs. It also features capabilities for image domain translation using paired or unpaired data and the ability to fill corrupted or missing image parts by analyzing surrounding visual context. The project covers ge
ArXiv | BibTex We develop a new approach for image inpainting that does a better job of reproducing filled regions exhibiting fine details inspired by our understanding of how artists work: lines first, color next. We propose a two-stage adversarial model EdgeConnect that comprises of an edge…
The main features of knazeri/edge-connect are: Generative Adversarial Networks, Image Inpainting, Image Inpainting Models, Shape Completion.
Open-source alternatives to knazeri/edge-connect include: advimman/lama — Lama is an image restoration framework and deep learning model designed for image inpainting and object removal. It… andreas128/repaint — Inpainting using Denoising Diffusion Probabilistic Models. kozistr/awesome-gans — Awesome-GANs is a curated resource list and research repository focused on the development and evaluation of… eriklindernoren/keras-gan — Keras-GAN is a collection of generative adversarial network implementations built with Keras for synthetic data… openai/glide-text2im — GLIDE is a generative model designed for text-to-image synthesis, image editing, and the contextual filling of masked… hybridgroup/gocv — GoCV is a computer vision library and Go language binding for OpenCV. It serves as an image processing toolkit and…