Invoke 4.2 is here: Control Layers let you apply regional guidance to specific areas of your image.

Invoke
9 May 202415:13

TLDRInvoke 4.2 introduces significant updates with the launch of the Community Edition and the addition of control layers in the commercial product. The new version features a redesigned interface with a new toolbar for easier workspace and viewer switching, and relocated control adapters and IP adapters to a dedicated 'Control Layers' tab. This update allows users to apply regional guidance to specific areas of an image, enhancing the precision of image generation. The 'Text to Image' tab has been revamped to include all core generation capabilities, and users can now add, delete, and manage layers with ease. The introduction of regional guidance layers is a major advancement, enabling users to influence the AI model's focus for certain elements within the composition. The video also discusses the interaction between global and regional prompts and offers tips on overcoming challenges associated with regional guidance. The creators of Invoke are eager to receive feedback and continue to improve the tool to better serve artists and creatives in their work with AI.

Takeaways

  • 🎉 Invoke 4.2 has been released with new features and a Community Edition.
  • 🖌️ Control Layers have been introduced to apply regional guidance to specific areas of an image.
  • 🔄 A new toolbar allows switching between a dedicated workspace and a viewer for image selection and transformations.
  • 📊 The interface has been updated with a new tab for control layers, moving control adapters and IP adapters from the settings panel.
  • 📈 The Text-to-Image tab has been updated to include all core generation capabilities.
  • 🎨 Regional guidance layers enable more precise control over where elements appear in the generated image.
  • 🔍 Users can visualize the placement of elements and add new regional controls to the generation process.
  • 📐 The ability to brush, erase, and use rectangle tools for regional masks provides detailed control over image composition.
  • ✅ Auto-negative feature automatically creates an inverted mask for regions, allowing for more nuanced control.
  • 🔄 Global and regional prompts interact to influence the AI model's focus and the final image output.
  • ⚙️ The global prompt has a significant impact on the overall image generation, guiding the broad understanding of the image's content.
  • 📝 Feedback and suggestions from users are encouraged to improve Invoke and its features continuously.

Q & A

  • What is the main feature introduced in Invoke 4.2?

    -The main feature introduced in Invoke 4.2 is the 'Control Layers' which allows users to apply regional guidance to specific areas of their image.

  • How can users switch between the workspace and the viewer in Invoke 4.2?

    -Users can switch between the workspace and the viewer by pressing the 'Z' key or by clicking the button in the top right corner of the interface.

  • What has been moved to the top left of the viewer in the new update?

    -The preview toggle and the metadata information have been moved to the top left of the viewer in the new update.

  • What is the new tab introduced for control adapters and IP adapters?

    -The new tab introduced for control adapters and IP adapters is called 'Control Layers'.

  • How has the 'Text to Image' tab been updated in Invoke 4.2?

    -The 'Text to Image' tab has been updated to 'Generation' and now includes all core generation capabilities such as text to image and image to image.

  • What are Regional Guidance Layers and how do they work?

    -Regional Guidance Layers are a major addition that allows users to prompt certain areas of the composition for specific elements. They work by influencing where the AI model focuses its attention for certain elements of the prompt.

  • How can users add or delete layers in the Control Layers tab?

    -Users can add layers by specifying the type of prompt (positive, negative, or image) and can delete layers individually or all at once by clicking the 'delete all' button.

  • What is the purpose of the 'Auto Negative' tag in the Regional guidance layers?

    -The 'Auto Negative' tag automatically creates an inverted mask and negatively prompts the opposite of that region, allowing users to control the generation in specific areas without affecting the rest of the image.

  • null

    -null

  • How does the global prompt interact with the regional prompt?

    -The global prompt provides an overall understanding of the image, while the regional prompt guides the AI model's attention to specific subjects or elements in certain areas of the image.

  • What is the role of the initial image in the image transformation process?

    -The initial image serves as a starting point for the image transformation. It can be optimized and adjusted to align with the desired output size and structural control of the generation.

  • How can users provide feedback or share their experiences with Invoke 4.2?

    -Users can reach out to the team on Discord to provide feedback or share their experiences with the new features in Invoke 4.2.

  • What are some of the user interface updates in Invoke 4.2?

    -Some of the user interface updates include a new toolbar on each tab, moving the control adapters and IP adapters into a new 'Control Layers' tab, and the addition of a preview toggle and metadata information to the top left of the viewer.

Outlines

00:00

🚀 Introducing Invoke 4.2: Interface Updates and New Features

The video introduces the Invoke 4.2 update, highlighting the release of the Community Edition and the addition of control layers in the commercial product. It emphasizes the excitement for future updates, which will bring more capabilities for control and Artistry within the Invoke tool. The interface has undergone changes, with elements moved around to accommodate new features. The video will discuss these changes and the new features added in the 4.2 update.

05:02

🎨 Exploring Control Layers and Regional Guidance in Invoke 4.2

The video delves into the new toolbar on each tab, which allows users to switch between a dedicated workspace and a viewer for image selection and transformation. The control adapters and IP adapters have been relocated to a new 'Control Layers' tab, streamlining the process of adding these elements. The 'Text to Image' tab has been updated to 'Generation,' incorporating all core generation capabilities. The video also introduces Regional guidance layers, which enable users to prompt specific areas of a composition for certain elements, providing more control over the AI model's focus. It discusses how global and regional prompts interact and the importance of aligning them for optimal results.

10:02

🖌️ Using Regional Masks and Controlling Image Composition

The video demonstrates the use of regional masks, including brushes, erasers, and rectangle tools for masking. It explains that while the process is not precise, it allows for significant control over certain regions of content. The video also discusses the 'Auto negative' feature, which automatically creates an inverted mask and negatively prompts the opposite of the selected region. It shows how to generate images with regional guidance and how to adjust the weight of terms to influence the output. The importance of a coherent global prompt that aligns with the desired material is emphasized, and the video concludes with a recommendation to experiment with the tools to understand their interaction and application in one's workflow.

15:05

📢 Final Thoughts and Invitation for Feedback on Invoke 4.2

The video concludes with a final test showcasing the control layers' ability to transform an image into a cat and a dog, emphasizing the potential of the tool for artists and creatives. It encourages viewers to subscribe and follow for the latest updates and to provide feedback or share experiences with the Invoke tool. The team is eager to hear how the tool is being used and how it can be improved to better serve the creative community.

Mindmap

Keywords

💡Invoke 4.2

Invoke 4.2 refers to the latest version of a software tool discussed in the video. It signifies a significant update that includes new features and improvements to existing functionalities, which are crucial for the users to understand the advancements in the tool.

💡Control Layers

Control Layers is a new feature in Invoke 4.2 that allows users to apply specific guidance to certain areas of an image. This feature is central to the video's theme as it enables more precise control over image generation and transformation, offering artists and creatives a higher level of control over their work.

💡Regional Guidance

Regional Guidance is a major addition to Invoke's capabilities, which allows users to direct the AI model's focus on specific parts of an image. It is exemplified in the video as a way to influence the aesthetic traits of certain elements within an image without affecting the rest, thus enhancing the creative process.

💡Global Prompt

A Global Prompt is a broad description or instruction given to the AI model that affects the entire image generation process. It is contrasted with Regional Prompts in the video, demonstrating how they interact to shape the final output of an image.

💡Viewer

The Viewer in the context of Invoke 4.2 is a user interface element that allows users to select gallery images, view them, and perform image-specific transformations. It is an essential tool for users to interact with and manipulate images within the software.

💡Workspace

The Workspace in Invoke 4.2 is a dedicated area for users to work on specific tasks related to the tabs in the software. It is highlighted in the video as a new feature that enhances the user's ability to organize and focus on particular aspects of image manipulation.

💡Control Adapters

Control Adapters are tools within Invoke 4.2 that users can utilize to add specific controls to their image generation process. They are mentioned as having been moved to a new tab called Control Layers, emphasizing their importance in customizing the AI's output.

💡Image Prompts

Image Prompts are inputs used in the image generation process that guide the AI model towards creating images based on the visual content provided. They are discussed in the video as a way to add visual elements to the generation process, either positively or negatively.

💡Text-to-Image

Text-to-Image is a core generation capability in Invoke 4.2 that allows users to create new images based on textual descriptions. It is part of the updated 'Text to Image' tab and is showcased as a way to generate images using control inputs.

💡Unified Canvas

The Unified Canvas is a feature that allows users to edit and composite generated images. It is mentioned in the context of making edits to images after they have been generated, providing users with a way to fine-tune their creations.

💡AI Model

The AI Model refers to the artificial intelligence system within Invoke 4.2 that processes the user's inputs and generates images accordingly. It is central to the video's narrative as it is the underlying technology that enables the creative process within the software.

Highlights

Invoke 4.2 introduces Control Layers, allowing for regional guidance on specific areas of an image.

The update includes a new toolbar for switching between workspaces and a viewer for image-specific transformations.

Control adapters and IP adapters have been moved to a new 'Control Layers' tab for easier access.

The 'Text to Image' tab has been updated to 'Generation', including all core generation capabilities.

Regional guidance layers enable specific prompts for certain areas of the composition, enhancing AI model focus.

Global and regional prompts can be used together for more coherent image generation.

Regional guidance layers can have positive, negative, and image prompts, offering fine control over visual aesthetics.

The AI model applies regional guidance with some freedom to maintain a globally coherent image.

Quality of life features include the ability to collapse layers for better visibility and change mask preview colors.

Auto-negative feature automatically creates an inverted mask for regions, allowing for nuanced control.

Aligned global and regional prompts prevent adjective bleed, ensuring distinct subjects in the image.

Global prompt weight significantly influences the output, even when regional guidance is applied.

Regional guidance can be used effectively with control adapters or compositional guidance for precise control.

The update offers new ways for artists and creatives to express themselves using AI tools.

Invoke continues to improve with each release, focusing on feedback from artists and creatives.

The team encourages users to reach out on Discord with feedback and share how they are using the tools.

More content and updates are coming, with an invitation to subscribe and follow for the latest information.