search

invoke-ai / InvokeAI

Invoke is a leading creative engine for Stable Diffusion models, empowering professionals, artists, and enthusiasts to generate and create visual media using the latest AI-driven technologies. The solution offers an industry leading WebUI, and serves as the foundation for multiple commercial products.
https://invoke-ai.github.io/InvokeAI/
Apache License 2.0
23.8k stars 2.45k forks source link

Add `ImagePanelLayoutInvocation` utility to facilitate In-Context LoRA workflows #7374

Closed RyanJDick closed 2 days ago

RyanJDick commented 6 days ago

Summary

This PR adds a ImagePanelLayoutInvocation that makes it easy to calculate the coordinates of panels within an image grid. This utility invocation is intended to make it easier to experiment with In-Context LoRA workflows.

Example 1: Reference-Based In-Context LoRA - Reference Logo Consistency

Instructions:

  1. Install this LoRA: https://huggingface.co/ali-vilab/In-Context-LoRA/blob/main/visual-identity-design.safetensors
  2. Load the sample workflow: flux_icl_single_reference_image.json
  3. Run.

Workflow

image

Sample output (cherry-picked, full grid): image

Sample output (cherry-picked, cropped): image

Example 2: Reference-Free In-Context LoRA - Character Consistency

Instructions:

  1. Install this LoRA: https://huggingface.co/ali-vilab/In-Context-LoRA/blob/main/film-storyboard.safetensors
  2. Load the sample workflow: flux_icl_reference_free.json
  3. Run.

Sample output: image

QA Instructions

Tested via the example workflows shown above.

Merge Plan

No special instructions.

Checklist