So, I will start by dropping a few keywords here, and what they are about, that will probably help you start, I am compiling this as a guideline on where to start
- VQGAN (Vector Quantized Generative Adversarial Network / neural network) : The software that generates the image
- CLIP (Contrastive Language-Image Pre-training / neural network) : Software to influence a generated image based on input text (User prompt)
- VQGAN+CLIP : Two neural network pieces of software that work in tandem.
- CLIP-Guided-Diffusion: A technique for doing text-to-image synthesis cheaply using pre-trained CLIP and diffusion models.
- Google colab notebook: A tool made by google where you can run python code and utilize google’s GPUs, both paid and free exist
Superior in both Lossless compression, and Lossy compression, webp is the new image format by google
Already supported by all web browsers *(that i have tested it with), webP is indeed a promising format, so let us get to compressing our images
I have a big bunch of bitmaps that my scanner spits out (To avoid lossy jpeg compression the scanner’s driver produces), and i need them converted to lossless webp to save space (the first image I compressed went from 552MB bitmap to 183), that is 33% of the original size
So, under linux, this is how i would convert all BMPs into webp images, I think it is exectly the same on windows
on the command line, the command for compressing one image looks like
cwebp -lossless 00.bmp -o 00.webp
Now, the next step is to run them in a batch, copy the following text into a file and name it with the extension