Giter Site home page Giter Site logo

Comments (5)

Svyaga avatar Svyaga commented on July 19, 2024

Forgot to mention that there was a similar problem in ComfyUI and it was fixed later:
comfyanonymous/ComfyUI#1363

And here's how it was fixed:
comfyanonymous/ComfyUI@bed116a

from invokeai.

Svyaga avatar Svyaga commented on July 19, 2024

Seems to be the same problem as in ComfyUI, here are some examples of misaligned pixels in initial generation of 888x1184 image:
Photoshop_TEbnbvMqYM
Photoshop_vl9gAr8HJH
I also should add that this happens on all sides of the borders, but it's more visible on bottom and right borders, especially right one.

from invokeai.

dunkeroni avatar dunkeroni commented on July 19, 2024

The first example does look like an issue I've seen when denoising padded latents, which may be the case if your region dimensions are not multiples of 64. Reflect padding instead of constants usually worked best in my experiments.

Your second example confuses me, because pixels far outside of your mask area are being affected. That looks like VAE nondeterminism to me, but the canvas paste-back operation shouldn't allow that. What is your Mask Blur setting at? Seems like it might be way too high.

from invokeai.

Svyaga avatar Svyaga commented on July 19, 2024

The first example does look like an issue I've seen when denoising padded latents, which may be the case if your region dimensions are not multiples of 64. Reflect padding instead of constants usually worked best in my experiments.

Your second example confuses me, because pixels far outside of your mask area are being affected. That looks like VAE nondeterminism to me, but the canvas paste-back operation shouldn't allow that. What is your Mask Blur setting at? Seems like it might be way too high.

  1. Dimensions could be the case for tiled img2img, but not for canvas and especially not the initial generation. I only use automatic dimensions with upscale to 1024x1024 (or similar dimensions for other aspect ratios). I mean the upscale if the selected box is smaller than model preffered dimensions ("Scale before processing" option).

Here's 1024x1024 image without any img2img applied:
a3f72176-ea9d-4c01-9332-d417c30cb4d9
You can notice the misaligned pixels there:
Photoshop_mij50rVRgx

  1. These are my settings for compositing:
    1111
    chrome_nREU2Zpd8u

from invokeai.

Svyaga avatar Svyaga commented on July 19, 2024

Info by dunkeroni:
https://discord.com/channels/1020123559063990373/1240001897814036562

from invokeai.

Related Issues (20)

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.