BeautifAI Image Upscaler! ft. MidJourneyMan & SuperBeast.ai // Civitai Guest Stream
TLDRIn a recent Twitch stream, Tyler from Civitai hosted a guest creator session with Julian (MidJourneyMan) and Brad (SuperBeast.ai), where they showcased an open-source image upscaling tool they've been working on for over a month. The tool, which can compete with paid upscalers, utilizes AI to enhance images and is designed for those looking to upscale without losing detail or clarity. The workflow includes options for two passes for further enhancement and resolution increase, and it's customizable, allowing users to adjust settings like the 'freedom' parameter for controlling the degree of image change and detail enhancement. The stream also covered tips for using the tool effectively, such as editing initial images to prevent artifacts, experimenting with different SDXL models, and using the seam fix mode for cleaner images. The presenters emphasized the importance of good prompts and clean inputs for optimal results and provided a link to the workflow for viewers to try for themselves.
Takeaways
- 🎨 Tyler, the host, introduces a guest stream featuring Julian (MidJourneyMan) and Brad (SuperBeast.ai), who are showcasing an open-source image upscaling tool they've been working on.
- 🔍 The upscaling tool is designed to compete with paid upscalers and is the result of over a month of collaboration and numerous iterations by Julian and Brad.
- 📈 The workflow includes a 'Freedom' parameter that allows users to control the degree of image enhancement and the introduction of new details.
- 🚀 The tool utilizes AI art techniques and is particularly useful for those looking to upscale images beyond basic capabilities while maintaining detail and clarity.
- ⚙️ The process involves two passes, with the second pass further enhancing the image and potentially increasing the resolution to 3 to 4K.
- 🖼️ Users can experiment with different SDXL models and prompts to achieve desired styles and effects, making the tool highly customizable.
- 💡 The stream highlights the importance of good initial images and prompts, as these factors significantly influence the output quality.
- 📉 The tool includes options for color matching, HDR effects, sharpening, and grain addition, all of which can be adjusted to refine the final image.
- 🤖 Julian and Brad discuss their personal experiences and preferences when using the tool, emphasizing the value of experimentation and iteration.
- 🧩 They also touch on the potential for editing the upscaled images in Photoshop for further refinement and detail correction.
- 🔗 The workflow is available for download on Civitai, and the creators encourage users to share their results and tag them on social media.
Q & A
What is the title of the stream?
-The title of the stream is 'BeautifAI Image Upscaler! ft. MidJourneyMan & SuperBeast.ai // Civitai Guest Stream'.
Who are the special guests featured in the stream?
-The special guests featured in the stream are Julian, also known as MidJourneyMan, and Brad, also known as SuperBeast.
What is the main topic of discussion in the stream?
-The main topic of discussion is an open-source image upscaling tool that Julian and Brad have been working on, which they believe can compete with many paid upscaling tools.
How long have Julian and Brad been working on the upscaling tool?
-Julian and Brad have been working on the upscaling tool for over a month, with around 12 or 13 versions and iterations.
What is the workflow's approach to upscaling images?
-The workflow combines various nodes and tools, including Ultimate Upscale and Control Nets, to maintain the structure of the image while upscaling. It also introduces a 'freedom' parameter to control the level of detail enhancement.
How can viewers find and download the workflow?
-Viewers can find and download the workflow by visiting civi.tha.com, searching for 'beyy', and following the provided link.
What is the significance of the 'freedom' parameter in the upscaling process?
-The 'freedom' parameter allows users to control the level of change and detail enhancement in the upscaled image. A higher value adds more details but can also introduce artifacts if set too high.
What are the system requirements to run the workflow?
-The workflow can run on systems with at least 8GB of VRAM, but for better performance and to handle larger image sizes, a system with 20GB or more is recommended.
How does the workflow handle multiple passes for upscaling?
-The workflow allows for two passes of upscaling. The first pass provides an initial upscale, and the second pass further enhances the image and increases the resolution, resulting in a more detailed final image.
What are some tips for using the upscaling tool effectively?
-Some tips include editing the original image to remove potential sources of artifacts, experimenting with different SDXL models and Lora prompts, and adjusting the 'freedom' parameter to balance detail enhancement with maintaining the original image structure.
How can users share their results and get feedback on their upscaled images?
-Users are encouraged to post their upscaled images on social media, tagging MidJourneyMan and SuperBeast to share their results and receive feedback.
Outlines
🎥 Introduction to the Stream and Guests
The paragraph introduces Tyler, the host of the Friday guest creator stream, who expresses excitement for the special guests, Julian (mid journeyman) and Brad (super beast). They discuss their experience with AI art and upscaling tools, and how they've been working on a workflow that can compete with paid upscalers. Julian and Brad share their journey with AI art and the challenges they faced, emphasizing their shift from MidJourney to Comy UI and their focus on enhancing details in their artworks.
📂 Downloading and Unpacking the Workflow
The paragraph explains the process of downloading the workflow from a provided link and extracting its contents. Julian guides the viewers on how to set up the workflow in Comy UI, emphasizing the importance of the Json and PNG files. They discuss the workflow's structure, the significance of the control nodes, and the benefits of using Ultimate Upscale and TT Planet Tyler for upscaling.
🎨 Upscaling Process and Parameters
In this paragraph, Julian and Brad delve into the specifics of the upscaling process, discussing the importance of the freedom value parameter and how it affects the upscaling outcome. They explain the workflow's design, from left to right, covering the image loading section, control nets, filters, and enhancement processes. The paragraph highlights the ability to experiment with different settings and models to achieve desired results.
🖌️ Enhancing Images and Freedom Parameter
The focus of this paragraph is on the enhancement of images through the freedom parameter. Julian and Brad discuss the impact of the parameter on detail enhancement and the introduction of new details. They also cover the use of color matching, filters, and sharpening tools. The paragraph emphasizes the importance of striking a balance between the freedom parameter and the original image to avoid artifacts and achieve a high-quality upscale.
🔍 Reviewing Results and Adjusting Workflow
The paragraph involves reviewing the upscaling results and discussing the workflow's performance. Julian and Brad provide insights into the mathematical values and control net strengths, and how they affect the upscaling. They also address the issue of artifacts and provide tips on adjusting the workflow, such as editing the original image and using different upscale models and seeds for varied results.
🌐 Tips, Tricks, and Final Thoughts
In the final paragraph, Julian and Brad share additional tips and tricks for using the workflow effectively. They discuss the importance of editing, the use of luras and prompts for style control, and the potential for using different models for various styles. The paragraph concludes with a discussion on the flexibility of the workflow, the importance of experimentation, and the potential for achieving high-quality results with the right combination of settings and models.
🤝 Acknowledgments and Closing Remarks
The paragraph wraps up the stream with acknowledgments and closing remarks. Tyler expresses gratitude to Julian and Brad for their contributions and shares links to the workflow and their social media profiles. He encourages viewers to experiment with the workflow and share their creations, emphasizing the value of community and collaboration in the AI art space. The paragraph ends with a preview of upcoming streams and a farewell to the viewers.
Mindmap
Keywords
💡BeautifAI Image Upscaler
💡MidJourneyMan
💡SuperBeast.ai
💡Civitai Guest Stream
💡Open Source
💡Upscaling Tools
💡Workflow
💡Control Nets
💡Freedom Value
💡Artifacts
💡Stable Diffusion
Highlights
Tyler welcomes viewers to the Civitai Friday guest Creator stream featuring special guests Julian (MidJourneyMan) and Brad (SuperBeast.ai).
Julian and Brad have been working on an open-source image upscaling tool that can compete with paid upscalers.
The upscaling workflow they developed has been iterated over 12 or 13 times to refine the process.
The tool incorporates AI art techniques and stable diffusion models to enhance image details.
Julian shares his experience of transitioning from MidJourney to Comfy UI for detailed image creation.
The upscaling process involves two passes for higher resolution and detail enhancement.
The workflow includes a 'freedom' parameter that controls the level of detail enhancement and image change.
The upscaling tool retains the original image's structure through the use of control nets.
The tool allows for custom nodes and has a version 2 with additional features like a custom node created by Brad.
Julian and Brad demonstrate the upscaling process with a live example, showing the before and after results.
The workflow is designed to be user-friendly, with drag-and-drop functionality and clear instructions for installation.
The presenters discuss the importance of using different seeds for pass one and pass two to prevent detail burn-in.
They highlight the ability to adjust the upscaling factor for both passes to achieve desired image resolutions.
Julian and Brad showcase the use of filters, sharpening, and grain addition to finalize the upscaled images.
The stream includes a Q&A session where the presenters answer questions about the workflow and upscaling techniques.
The workflow is open for modification, allowing users to experiment with different settings and nodes to improve it.
The presenters encourage users to share their upscaled images on social media and tag the creators for feedback and sharing.
Tyler thanks Julian and Brad for their contribution and discusses the importance of community collaboration in AI art development.