Replies: 8 comments 6 replies
-
Hi, |
Beta Was this translation helpful? Give feedback.
-
yep i know the basic, as been doing since feb. unless i add like wearing black shoes, etc doesn't want to do anything more than belly up for the last month |
Beta Was this translation helpful? Give feedback.
-
strange as some times I change the ratios of image size and it doesnt change the output
Sent from Outlook for Android<https://aka.ms/AAb9ysg>
…________________________________
From: poor7 ***@***.***>
Sent: Monday, August 12, 2024 9:59:49 AM
To: lllyasviel/Fooocus ***@***.***>
Cc: James ***@***.***>; Author ***@***.***>
Subject: Re: [lllyasviel/Fooocus] still not getting full image of head to toe (Discussion #3486)
There is probably something wrong with your hints or model. What model are you using for generation? Can you provide some examples of hints and the resulting images?
—
Reply to this email directly, view it on GitHub<#3486 (reply in thread)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/AIPK7J5CJXLTRCSOBLDIBZ3ZRB2QLAVCNFSM6AAAAABMJ4MXFCVHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTAMZRGE3DAMY>.
You are receiving this because you authored the thread.Message ID: ***@***.***>
|
Beta Was this translation helpful? Give feedback.
-
prompt set to 832x1216 and get should be a longer image changed to
set to 704x1408 ( should be a lot skinner pic ) ideals? |
Beta Was this translation helpful? Give feedback.
-
changed model to realisticStockPhoto_v20 |
Beta Was this translation helpful? Give feedback.
-
so changed a few bits, preset realistic, size 704x1408, got this ( looks more full body??) as soon as i add jessicaL for my lora i go to prompt jessicaL ,__ perspective__ , emotion, pose wearing a Clothing_Designers clothing-female , standing on beach in turkey, pencil eyeliner, DSLR, parted lips, pink lipstick, silver necklace with a cross on it, straw hat, Full body image, so this means my lora is over setting the image size for full body? |
Beta Was this translation helpful? Give feedback.
-
I have been advised this is more that my lora was not trained on full images and also foocus somehow is letting the lora over write my prompts. I moved to Sd this week to test and the same thing
Sent from Outlook for Android<https://aka.ms/AAb9ysg>
…________________________________
From: Rogala ***@***.***>
Sent: Sunday, August 18, 2024 9:20:38 PM
To: lllyasviel/Fooocus ***@***.***>
Cc: James ***@***.***>; Author ***@***.***>
Subject: Re: [lllyasviel/Fooocus] still not getting full image of head to toe (Discussion #3486)
Try this LoRA but initially it should be checked and adjusted to your character, its weight can vary from -2 to 2, and sometimes -1, 0, 1 only give normal results but in any ratio of the sides of the picture.<https://civitai.com/models/213307/envy-zoom-slider-xl-01?modelVersionId=240274>
—
Reply to this email directly, view it on GitHub<#3486 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/AIPK7J5YZ66JITOIOZ4GF7LZSD6ZNAVCNFSM6AAAAABMJ4MXFCVHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTAMZXGU3TSOA>.
You are receiving this because you authored the thread.Message ID: ***@***.***>
|
Beta Was this translation helpful? Give feedback.
-
hi all
trying for a full body image, tried in the prompt standing , head to toe image, medium shot etc and not getting any full person images?
Beta Was this translation helpful? Give feedback.
All reactions