We looked through thousands of GAN images. Choosing one was the hard part.
The model had finished generating. That did not mean we had a finished artwork. In Whispart's studio workflow, it meant we had a much more difficult question: which image would still be interesting after the first two seconds? We make images with a GAN model we call Mnemosyne . Our founder describe…
Read the full story at DEV Community — Machine Learning ↗
Timeline · 1 report
- 2026-09-24 09:29 · DEV Community — Machine Learning
We looked through thousands of GAN images. Choosing one was the hard part.