VQGAN+CLIP is a text-to-image model that generates images of variable size given a set of text prompts and some other parameters. Katherine Crowson, artist and mathematician wrote the Google Colab Notebook that combined VQGAN + CLIP. The notebook was originally in Spanish and has later been translated into English.
VQGAN+CLIP is a combination of two neural network architectures – VQGAN and CLIP – that allows you to train a neural network on anything from text to images. VQGAN involves convolutional neural networks traditionally used for images with Transformers.
CLIP is a model that was originally intended for searching for the best match to a description like “a dog playing the violin” among a number of images. The revolutionary thing about CLIP is that it is capable of zero-shot learning and performs exceptionally well on previously unseen datasets.
VQGAN chooses an image that is the best match for a given text, while CLIP chooses the best image for a particular word or phrase.
Further reading: Alexa Steinbrück
Ulupamir (meaning Great Pamir) is a village in Turkey of mostly Kyrgyz Turks who came from the Great Pamir and Little Pamir in the Wakhan, northern Afghanistan, in 1982. Pamir Turks first fled to Pakistan but they did not adjust well to the hot Pakistani climate and lost lives. With the invitation from Turkey they moved to a land near Van Erciş. This is a photo reportage about a culture on the verge of change.
Beauty of Patterns
Patterns are everywhere… As Fibonacci numbers… I was always admired with the beauty of it and wanted to create visual patterns created from simple geometric shapes. It is so interesting that how randomness can perfection.
I created more than 10k patterns for stock agencies and later decided to create them as #NFTs.
Stock Photography and Stock Vectors
Working with stock agencies provides me with a sense of freedom. I’m a member of several agencies, including Fotolia, Shutterstock, Alamy, Getty, and others, but I’m thinking about going exclusive. I’ve written a number of articles on the subject, which I’ll post here from time to time. However, my perspective on stock photography has evolved significantly over the years.