【Stable Diffusion】ポーズの呪文集プロンプト一覧を紹介！
TLDRThe video provides a comprehensive guide on using 'pose spells' to generate a variety of poses in stable diffusion AI. It emphasizes the importance of entering the correct pose spell to achieve the desired outcome, as stable diffusion may not produce poses without it. The video introduces various pose spells, such as 'standing, raise one leg' for a leg-up stance, 'fold-arms' for hiding hands, and 'hold the steering wheel of a car' for a driving pose. It also offers tips on image generation, like using aspect ratios to influence pose outcomes and adjusting the spell's influence with a numerical value (up to 1.3) to improve results without compromising quality. The video concludes with the suggestion to use 'open pose editor' for more complex or unique poses and encourages viewers to experiment with different spells to achieve their ideal image.
Takeaways
- 🔍 Use specific 'pose spells' to increase the variation and probability of generating desired poses in stable diffusion.
- 🧍 If you want a standing pose, use the 'standing' spell for better results.
- 🦵 To generate a pose with one leg raised, add 'standing, raise one leg' to the prompt.
- 🤗 For folded arms, use 'fold-arms' or 'crossed arms' to hide hands and improve image generation.
- 🤲 Adding 'arms behind back' or 'hand between legs' can help in creating poses where hands are less visible.
- 🚗 For poses involving objects like steering wheels, use descriptive spells like 'hold the steering wheel of a car'.
- 🤚 When aiming for a pose with hands reaching out, be aware it can be challenging to generate due to finger details.
- 🙌 'Arms raised in the air' is a common pose that can be used for various expressive gestures.
- 👶 For child sitting poses, use 'sit' or 'sitting', and consider vertical image sizes for balance.
- 🏃♂️ To generate walking or running poses, use 'walking' or 'running', and be patient as it may take several attempts.
- 🛏️ For lying down or sleeping poses, adjust the aspect ratio to vertical for supine or horizontal for side sleep.
Q & A
What is the importance of using 'pose spells' in Stable Diffusion?
-Pose spells in Stable Diffusion are crucial for generating specific poses in images. Without the correct pose spell, the desired pose may not appear, or it may take a long time to generate by chance. Using the right spell increases the probability of achieving the intended pose.
How can one increase the probability of generating a standing pose in Stable Diffusion?
-To increase the probability of generating a standing pose, one should use the 'standing' spell. Additionally, for a pose with one leg up, the spell 'standing, raise one leg' can be added.
Why might someone want to hide the hands in an AI-generated image?
-Hiding the hands can be beneficial because AI illustrations often struggle to create detailed and beautiful fingers. By hiding the hands, it becomes easier to produce high-quality image generation works.
What is a recommended spell for those who want to depict a character with their arms behind their back?
-The recommended spell for depicting a character with their arms behind their back is 'arms behind back'. This is also a good method for hiding the hands.
How can one generate an image of a character sitting on a chair?
-To generate an image of a character sitting on a chair, one can use the spell 'sit' or 'sitting'. It's also recommended to use a vertically long image size for a well-balanced result.
What aspect ratio is recommended for generating images of typical poses?
-A vertical aspect ratio is recommended for generating images of typical poses, as it tends to be easier to create standing or sitting poses in this format.
How can one adjust the influence of a spell in Stable Diffusion?
-To adjust the influence of a spell, one can add a numerical value after the spell, following a colon, like in 'run1.3'. The recommended value is up to 1.3 to avoid deteriorating the overall image quality.
Why is it suggested to put the pose spell at the beginning of the prompt in Stable Diffusion?
-Placing the pose spell at the beginning of the prompt increases the likelihood of generating the desired pose because the prompts in Stable Diffusion's web UI tend to prioritize spells that are listed first.
What is the 'open pose editor' and how can it be used to generate an ideal pose?
-The 'open pose editor' is a tool that allows for more precise specification of poses that may not be covered by the standard spells. It can be used to generate poses that are not introduced in the standard spell collection or poses that are difficult to generate.
How can changing the aspect ratio help in generating poses that are difficult to achieve?
-Changing the aspect ratio to vertical, square, or landscape can influence the type of poses that are generated. For example, a vertical ratio is better for standing or sitting poses, while a square or landscape ratio might be more suitable for poses where the character is sitting on the ground or lying face down.
What is the recommended approach when a specific pose is not being generated as desired?
-If a specific pose is not being generated as desired, it is recommended to change the aspect ratio, adjust the spell's influence by adding a value up to 1.3 after the spell, and prioritize the pose spell by placing it at the beginning of the prompt.
Outlines
🎭 Introduction to Pose Spells for Image Generation
The first paragraph introduces the concept of 'pose spells' in the context of stable diffusion image generation. It explains that finding the right pose spell can be challenging and time-consuming, but the video aims to help by providing a collection of these spells. The paragraph covers various pose spells such as 'standing', 'standing, raise one leg', 'fold-arms', 'arms behind back', 'hand between legs', 'put hands hip', 'one hand on hip', 'own hands together', 'hold the steering wheel of a car', 'reaching out', 'arms raised in the air', 'piece sign', 'put up index finger', 'sit', 'sitting', 'lie down', 'sleep', 'closed eyes', 'lie face down', 'crossed leg', 'kneel', 'own hugging feet', 'all fours', 'wariza', and 'seiza'. It also discusses tips for hiding hands and the importance of aspect ratios in generating poses, with a focus on vertical, square, and landscape orientations.
📐 Tips for Generating Ideal Poses
The second paragraph provides detailed tips for successfully generating the desired pose in image generation software. It suggests changing the aspect ratio to a vertical, square, or landscape format to influence the type of pose that is generated. The paragraph also recommends using a numerical value after the pose spell to adjust its influence, with a suggested maximum of 1.3 to avoid quality deterioration. It emphasizes the importance of prioritizing certain spells at the beginning of the prompt to increase the likelihood of generating the intended pose. The paragraph concludes with a reminder that generating a specific pose may require multiple attempts due to the probabilistic nature of image generation and suggests using 'open pose' or 'open pose editor' for more control over the process.
🔄 Summary and Conclusion
The final paragraph summarizes the key points discussed in the video. It reiterates the effectiveness of changing the aspect ratio when attempting to generate a specific pose that is not appearing. It also highlights the use of a numerical value to strengthen the influence of a pose spell and the importance of placing the desired pose spell at the beginning of the prompt for higher priority in generation. The paragraph encourages viewers to refer to the spells introduced and to experiment with generating their ideal poses. It concludes by inviting viewers to subscribe to the channel for more content on generated AI and thanking them for watching.
Mindmap
Keywords
💡Stable Diffusion
💡Pose Spells
💡Image Generation
💡Aspect Ratio
💡Standing Spell
💡Folding Arms
💡Hands on Hips
💡Running
💡Lying Down
💡Sleeping
💡Open Pose Editor
Highlights
Stable diffusion often requires entering the correct pose spell to generate a desired pose.
Referring to a collection of pose spells can help quickly finish the ideal pose.
Use the standing spell to generate a standing picture of a person.
Adding 'standing, raise one leg' can generate a pose with one leg up.
Combining spells, like 'fold-arms' or 'crossed arms', can help hide hands and improve image generation.
Adding 'arms behind back' is a recommended method for those who want to conceal their hands.
Including 'hand between legs' and 'sit' can enhance the generation of sitting poses.
Using 'put hands hip' or 'one hand on hip' can generate poses with hands on the hips.
Adding 'own hands together' can create a pose with hands clasped, though it's challenging to generate natural-looking fingers.
Entering 'hold the steering wheel of a car' can generate poses with the character driving.
Using 'reaching out' can create poses with hands extended, though it's difficult due to finger details.
Artistic poses like 'arms raised in the air' can be generated by using the corresponding spell.
Adding 'piece sign' can be used when creating a piece, though it may not bend the thumb.
To generate a sitting pose, use 'sit' or 'sitting', and consider a vertically long image size for balance.
Inserting 'walking' or 'running' can generate poses of characters in motion, though it may require multiple attempts.
Adding 'jumping' can create dynamic poses, but the probability of a beautiful generation is low.
For a supine pose, use 'lie down' and choose a vertical aspect ratio for better results.
To generate a sleeping pose, use 'sleep' and consider 'sleep, closed eyes' for a more realistic image.
Changing the aspect ratio to vertical, square, or landscape can influence the generation of different poses.
Adjusting the spell's influence with a numerical value (e.g., run1.3) can help achieve the desired pose.
Placing the pose spell at the beginning of the prompt can increase the likelihood of its generation.
Using 'open pose editor' can help generate poses not covered in the introduced spells.
