thygate / stable-diffusion-webui-depthmap-script

High Resolution Depth Maps for Stable Diffusion WebUI
MIT License
1.72k stars 159 forks source link

[Feature request] Automatically reuse already generated depthmaps #255

Open donbrasco01 opened 1 year ago

donbrasco01 commented 1 year ago

Discussed in https://github.com/thygate/stable-diffusion-webui-depthmap-script/discussions/248

Originally posted by **donbrasco01** June 6, 2023 Hello people, i´m wondering if the depthmap files in the output directory need to have a specific name to be seen by the script, because its always trying to create new depthmaps with the selected model and not using the already existing depthmaps. Thanks in advance !
semjon00 commented 1 year ago

I like this idea. This may be done in future, after https://github.com/thygate/stable-diffusion-webui-depthmap-script/pull/278 is complete. Implementing this fully would be quite hard - so maybe not a priority, but a moderately easy solution would be to keep the last generated depthmap and reuse it if the model, image and the depthmap generation parameters are the same. For example, this would be useful when trying out different stereo generation options.

semjon00 commented 1 year ago

i´m wondering if the depthmap files in the output directory need to have a specific name to be seen by the script, because its always trying to create new depthmaps with the selected model and not using the already existing depthmaps.

@donbrasco01 Sorry, I guess I mis-interpreted your original question. Please try the Batch from Directory option with "Skip generation..." enabled.

donbrasco01 commented 1 year ago

Thank you for your reply! I just figured it out .... the externally generated depthmaps need to have the "original-filename"+"-0000" and they have no be PNG. Its working now exactly as it supposed to be!

Thank you again