Open-source project that creates or edits images with generative AI and text-based controls.
CurveLab is a toolbox implementing the Fast Discrete Curvelet Transform in Matlab and C++, currently at version 2.1.2. Academics at not-for-profit organizations may download the source code for non-commercial research after registering and signing a license agreement. Commercial use and use by for-profit organizations require a commercial license from the Caltech Office of Technology Transfer, which includes a free trial period.
A blog post explaining why upscaling AI-generated images matters in image-to-image workflows. It describes how diffusion models apply and remove noise, and how higher-resolution reference images give the model more defined details. The post compares a low-resolution and upscaled example using Stable Diffusion's ControlNet with the canny model, and mentions upscaling options including ImgUpscaler, TopazLabs and Upscale Media.
Hito Steyerl, Mean Images, NLR 140/141, March–June 2023 creates or edits images with generative AI and text-based controls.
Computer Vision Art Gallery is a site for the CVPR 2021 art gallery, held virtually due to COVID-19. It lists a winner, shortlisted works and all accepted works, each credited to artists with titles and years.
Partially Derivative is a podcast hosted by Chris and Vidya, described as "Car Talk for the data community." Episodes discuss data science, artificial intelligence, deep learning, p-values and related topics. The site notes that the last episode was released for now in September 2017.
PyTTI is a toolkit that facilitates image generation, animation, and manipulation through processes described as a human artist collaborating with AI assistants. The project states that users do not need to be deep learning experts or know coding to use the tools.
Community notes
Be the first to share how this project worked for you.
Sign in with a verified account to leave a review or join the discussion.
Sign in to reviewKeep exploring
CurveLab is a toolbox implementing the Fast Discrete Curvelet Transform in Matlab and C++, currently at version 2.1.2. Academics at not-for-profit organizations may download the source code for non-commercial research after registering and signing a license agreement. Commercial use and use by for-profit organizations require a commercial license from the Caltech Office of Technology Transfer, which includes a free trial period.
A blog post explaining why upscaling AI-generated images matters in image-to-image workflows. It describes how diffusion models apply and remove noise, and how higher-resolution reference images give the model more defined details. The post compares a low-resolution and upscaled example using Stable Diffusion's ControlNet with the canny model, and mentions upscaling options including ImgUpscaler, TopazLabs and Upscale Media.
Hito Steyerl, Mean Images, NLR 140/141, March–June 2023 creates or edits images with generative AI and text-based controls.
Computer Vision Art Gallery is a site for the CVPR 2021 art gallery, held virtually due to COVID-19. It lists a winner, shortlisted works and all accepted works, each credited to artists with titles and years.
Partially Derivative is a podcast hosted by Chris and Vidya, described as "Car Talk for the data community." Episodes discuss data science, artificial intelligence, deep learning, p-values and related topics. The site notes that the last episode was released for now in September 2017.
PyTTI is a toolkit that facilitates image generation, animation, and manipulation through processes described as a human artist collaborating with AI assistants. The project states that users do not need to be deep learning experts or know coding to use the tools.
