Compressing Images with Stable Diffusion

Re:Compressing Images with Stable Diffusion by WheelsAtLarge on Thursday September 01, 2022 @12:38AM (#32672067)

You would think but there seems to be some kind of randomization added to the image creation. You can use the same prompt over and over again and get different images.

[ Parent | on HN ]

Re:Compressing Images with Stable Diffusion by bitshiftfaced on Thursday September 01, 2022 @12:39AM (#32672073)

When I use the same seed and same prompt, I get the same image.

[ Parent | on HN ]

Re:Compressing Images with Stable Diffusion by nextaccountic on Thursday September 01, 2022 @01:57AM (#32672531)

Same seed, same prompt and same hardware perhaps, if the GPU concurrency doesn't introduce nondeterminism.

Otherwise you might have the same problem of this: https://github.com/EmbarkStudios/texture-synthesis#notes

> When using multiple threads for generation, the output image is not guaranteed to be deterministic with the same inputs. To have 100% determinism, you must use a thread count of one, which can by done via

> CLI - texture-synthesis --threads 1

> API - SessionBuilder::max_thread_count(1)

[ Parent | on HN ]

Re:Compressing Images with Stable Diffusion by astrange on Thursday September 01, 2022 @12:54AM (#32672158)

Compression algorithms using floating point have always been flaky though. Need a fixed-point image model.

[ Parent | on HN ]

Re:Compressing Images with Stable Diffusion by WheelsAtLarge on Thursday September 01, 2022 @01:35AM (#32672401)

You are correct. My mistake.

[ Parent | on HN ]

Re:Compressing Images with Stable Diffusion by Agentlien on Thursday September 01, 2022 @12:59AM (#32672183)

I've played a lot with this and I get the same image using the same prompt and seed.

I've also played a lot with slight variations on the same prompt and it's really fascinating seeing how similar the images are outside of the specific detail changed.

I would say this diffusion model is very... stable.

[ Parent | on HN ]