Midjourney Panning + Dreamshaper Version 7

Olivio Sarikas
4 Jul 202311:20

TLDRIn this video, Team Nostromo introduces the new panning function of Midjourney and the capabilities of Dream Shaper Version 7. The panning feature allows for extensive image expansion in any direction, enhancing the cyberpunk cityscape with a realistic build-up. Dream Shaper 7 is highlighted for its versatility in creating photorealistic, digital art, and anime images. The video demonstrates the model's proficiency with realistic styles, particularly in rendering hair and armor, and provides tips on using negative prompts to refine outputs. It also covers the importance of the CLIP Skip 2 feature and how to implement it for better image quality. The presenter shares various prompts and negative prompts to achieve high-quality results, including detailed skin and sharpness enhancements. The video concludes with a showcase of different artistic styles created with the model and invites viewers to suggest the next model to be featured.

Takeaways

  • 🎨 **Midjourney Panning Feature**: The new panning function in Midjourney allows users to extend their images as wide as desired, enhancing the scene with additional details that make sense in the context of the image.
  • 📈 **Image Upscaling**: To use the panning feature, one must first create an image and then upscale it using the 'U' buttons. After upscaling, directional arrows appear for extending the image in the chosen direction.
  • ➡️ **Directional Extension**: Once a direction is chosen for image extension, the same choice will be presented next time, limiting the user to extending in the previously selected horizontal direction.
  • 🔧 **Remix Mode Activation**: If the panning option does not appear, users should access Midjourney settings and activate the remix mode to enable the feature.
  • 🔄 **Dream Shaper Version 7**: This version of the model can create photorealistic, digital art, and anime images, showcasing a variety of styles with high-quality outputs.
  • 🖌️ **Image Quality**: The model excels in rendering details like hair and armor, providing photorealistic results that are very good in terms of texture and reflection.
  • 🚫 **Negative Prompts**: The use of negative prompts such as 'bad dream' and 'unrealistic dream' can help improve image quality by guiding the model to avoid certain undesirable outcomes.
  • 📁 **File Management**: Model files should be downloaded into the 'stable diffusion' folder within 'models', while negative embeddings go into the 'embeddings' folder for proper functionality.
  • 📊 **CFG Scale and CLIP Skip**: The model utilizes the CLIP Skip 2 feature, which is crucial for achieving the correct output. Users can adjust settings such as the CFG scale and choose the right CLIP Skip version.
  • 🔍 **Image Sharpness**: The model may struggle with sharpness in upscaled images. Using terms like 'full sharp' and detail-oriented tweaks can enhance the clarity and resolution of the final output.
  • 🌟 **Detail Tweaker Tools**: Tools like the 'Laura funic stress' and 'One Button Prompt Extension' can be used to add more details and improve the overall quality of the generated images.
  • 🖋️ **Artistic Styles**: The model supports various artistic styles, offering users the flexibility to create images that are not only realistic but also stylized according to different artistic preferences.

Q & A

  • What is the new panning function in Midjourney?

    -The new panning function in Midjourney allows users to extend their image in any direction as wide as they want, adding parts to the scene and building up the cityscape in a way that makes sense.

  • How do you upscale an image in Midjourney?

    -To upscale an image, you need to create an image and then click on one of the 'U' buttons. After the image is upscaled, you can use the arrows pointing left, right, up, and down to extend the image in the chosen direction.

  • What should you do if the panning option does not show up?

    -If the panning option does not show up, you should go to Midjourney, type '/settings' and activate the remix mode in the settings window.

  • What are the capabilities of Dream Shaper Version 7?

    -Dream Shaper Version 7 is capable of producing photorealistic images, digital art, and even anime styles, offering a wide range of styles executed in a well-polished manner.

  • How can you improve the image quality using negative embeddings?

    -Negative embeddings, such as 'bad dream' and 'unrealistic dream', are specifically made for the Dream Shaper model to improve image quality. They can be added to the negative prompt field in the textual inversion tab to enhance the output.

  • Where should you download the model file and negative embeddings?

    -The model file should be downloaded into the 'automatic 1111' folder, then into the 'models' folder, and finally into the 'stable diffusion' folder. The negative embeddings should be downloaded into the 'embeddings' folder within the 'automatic 1111' directory.

  • How do you choose the Clip Skip and SDVAE options in the settings?

    -In the settings, navigate to the 'user interface' section and scroll down to the 'quick setting list'. You can then select 'SDVAE' and 'clip stop at last layers' from the list provided when you click inside the field.

  • What is the special trick mentioned for improving the quality of upscaled images?

    -To improve the quality of upscaled images, the video suggests using the term 'full sharp' enclosed in double round brackets, which has a high impact on the image. Additionally, using 'highly detailed' and the 'Laura at detail' tweaker can further enhance the image's sharpness and detail.

  • How does the model handle photorealistic images of hair?

    -The model is particularly good with photorealistic images of hair, where each hair strand is visible, contributing to the overall quality and realism of the image.

  • What is the issue with the upscaled images in terms of sharpness?

    -The upscaled images sometimes have a problem with getting completely sharp and crisp details, especially when viewed at a larger size or when pixel peaking.

  • What is the recommended resolution for creating high-quality images with Dream Shaper Version 7?

    -The recommended resolution for creating high-quality images with Dream Shaper Version 7 is 1024 by 1536, as suggested by the model's page.

  • How can the 'One Button Prompt Extension' help in creating images?

    -The 'One Button Prompt Extension' can suggest amazing prompts that can be used to create beautiful images. It is a useful tool for generating creative and detailed images without manually crafting the prompts.

Outlines

00:00

🎨 Introduction to Mid-Journey Panning and Dream Shaper 7

The video begins with an introduction to the new panning function in Mid-Journey, a tool that allows users to extend their image in any direction for a customizable and expansive view. The host, Team Nostromo, demonstrates the feature's ease of use and its ability to enhance scenes by adding parts that fit seamlessly into the existing image. The video also introduces Dream Shaper version 7, a versatile model capable of generating photorealistic, digital art, and anime-style images. The host emphasizes the importance of exploring all settings and provides a guide on how to activate the remix mode for optimal use. Negative prompts are introduced as a method to refine the model's output, with specific negative embeddings designed for the Dream Shaper model to improve image quality. The host guides viewers on where to download and install the model file and negative embeddings for use in their Automatic1111 software.

05:00

📈 Customizing Settings and Enhancing Image Quality

The host explains how to customize settings within the Automatic1111 interface, specifically showing how to add negative prompts and select Clip Skip and SDVAE options. The video provides examples of image outputs using the Dream Shaper 7 model, highlighting the model's proficiency with photorealistic images, particularly in rendering hair and chainmail with impressive detail and realism. However, the host also acknowledges that the model sometimes struggles with maintaining sharpness during upscaling. To address this, the host shares a trick to improve image quality and demonstrates the effectiveness of using detailed prompts and the Laura detail tweaker to enhance the clarity and detail of the final images.

10:02

🌟 Exploring Artistic Styles and Closing Remarks

The video concludes with a showcase of additional images created using the Dream Shaper 7 model in various artistic styles, emphasizing the model's versatility and the high quality of the results. The host encourages viewers to share their thoughts on the video and to suggest models for future videos. Before ending, the host also hints at additional content available for viewers to explore and encourages them to like the video and look forward to future encounters.

Mindmap

Keywords

Mid-journey panning

Mid-journey panning refers to a new function in the video that allows for the creation of a panoramic image by extending the view in a chosen direction, such as left or right. This feature is significant as it enables the user to capture a wider scene, adding depth and detail to the image, especially in the context of a cyberpunk cityscape depicted in the video.

Dream Shaper Version 7

Dream Shaper Version 7 is a new version of a model discussed in the video that is capable of generating photorealistic, digital art, and even anime images. It is highlighted for its versatility and the quality of its outputs across different styles. The video demonstrates how this model can be used to create high-quality images by adjusting various settings and using specific prompts.

Upscaling

Upscaling in the context of the video refers to the process of increasing the resolution of an image, which is done by clicking on one of the 'U' buttons as mentioned in the script. Upscaling is crucial for enhancing the detail and clarity of the image, and it is a step used before extending the image using the mid-journey panning feature.

Negative Prompt

A negative prompt is a term used to describe specific instructions given to the model to avoid generating certain unwanted elements or styles in the output image. In the video, 'bad dream' and 'unrealistic dream' are mentioned as negative prompts used to improve the image quality by guiding the model to produce more realistic and desired results.

Clip Skip

Clip Skip is a parameter used in the model that affects how the model processes the input to generate the output image. The video emphasizes the importance of using the correct version of Clip Skip, specifically Clip Skip 2, for the best results with the Dream Shaper model.

Photorealism

Photorealism is a style of art where the images are created to closely resemble real-life photographs. The video discusses the Dream Shaper model's capability to produce photorealistic images, highlighting the quality of details such as hair and armor in the generated images.

Digital Art

Digital art refers to artistic works or designs created using digital technology, often on computers. The video showcases the Dream Shaper model's ability to create digital art with various styles, demonstrating its flexibility and the high-quality output it can produce.

Anime Images

Anime images are a style of illustrations that originate from Japanese animation and comic styles. The video mentions the Dream Shaper model's ability to generate anime-style images, showcasing its versatility in producing different types of visual art.

CFG Scale

CFG Scale is a term likely referring to a setting or parameter within the model that controls the level of detail or configuration of the generated image. The video does not provide explicit details about what CFG Scale does, but it is presented as an important setting to adjust for optimal results.

One Button Prompt Extension

The One Button Prompt Extension is a tool or feature mentioned in the video that assists users by suggesting prompts for image generation. It is praised for its ability to recommend creative and effective prompts, simplifying the process of generating detailed and compelling images.

Detail Tweaker

A Detail Tweaker is a tool or setting that allows users to enhance the level of detail in the generated images. The video specifically mentions the 'Laura at Detail' tweaker as a way to add more detail to the images, making them sharper and more visually appealing.

Highlights

Introduction to Midjourney Panning and Dreamshaper Version 7 with Team Nostromo

New panning function allows for wide panoramic images of cyberpunk cityscapes

Easy-to-use interface for image upscaling and extending in any direction

Activation of remix mode in settings for additional features

Dreamshaper Version 7 capable of photorealism, digital art, and anime styles

Improved image quality with hair and armor details

Use of negative prompts to refine outputs and address model limitations

Downloading and installing model files and negative embeddings for improved results

Configuring the AUTOMATIC1111 for model and negative embeddings usage

Explanation of CLIP Skip and its importance for the model

Adding negative embeddings to prompts for better image quality

Customizing UI settings for CLIP Skip and SDVAE quick selection

Presentation of sample renders showcasing the model's capabilities

Tips for using detailed prompts to enhance image resolution and clarity

Techniques to improve sharpness and detail in upscaled images

Demonstration of the model's effectiveness with various artistic styles

Invitation for viewers to request models for future videos

End screen with additional content suggestions and a call to action for likes