How to make AI art of yourself with Midjourney for free
TLDRThe video introduces 'Mid-Journey', an AI tool that generates unique images from text prompts. Users can create detailed scenes, combine images, or even transform styles to anime. The AI offers four variations per prompt, which can be refined for better results. Tips for effective use include using a template for precision, re-rolling for distinct characters, specifying high detail for photorealism, adding camera perspectives, using 'niji' for anime style, and adjusting aspect ratios for different compositions. The video also promotes 'Pirate Tea', a natural energy drink.
Takeaways
- 🎨 Mid-journey is an AI that generates new images from text or image inputs.
- 🔄 You can combine text and images, or mash two images together for unique creations.
- 🆓 Mid-journey is a free service that can be accessed with a Discord account.
- 📝 To create an image, use the command '!imagine' followed by your prompt in the Discord channel.
- 🔄 Each prompt will generate four variations, and you can select the one you like best.
- 🖼️ Images can be upscaled, and you can include an image URL in your prompt by pasting it at the beginning.
- 🎨 For more control, use a template to give Mid-journey specific instructions.
- 👥 Creating distinct characters can be challenging, so re-rolling may be necessary.
- 📸 Adding details like 'Ultra HD photography' can enhance photorealistic outputs.
- 📐 Include camera perspectives (e.g., 'GoPro perspective') for a more dynamic result.
- 🌟 Use 'candid photo' for subjects engaged in activities rather than looking directly at the camera.
- 🎭 For anime-style images, add '--niji' to your prompt.
- 📈 Change the aspect ratio of images by adding '---AR' followed by the desired ratio.
Q & A
What is Mid-Journey and how does it function?
-Mid-Journey is an AI that generates new images based on the input provided to it. Users can create prompts using text, images, or a combination of both, and the AI will produce images accordingly.
How can a user get started with Mid-Journey?
-To use Mid-Journey, a user needs to sign up on their website using a Discord account. Once in the Discord channel, users can start creating prompts in the newbie rooms by typing 'Imagine' followed by their prompt.
What kind of prompts can be used with Mid-Journey?
-Prompts can be textual descriptions, such as 'Samurai painting in the style of Edward Monet', or combinations of text and images, or even a blend of two different images.
How does Mid-Journey handle multiple variations of an image?
-For each prompt, Mid-Journey provides four variations of the image. Users can select the one they like best by typing 'U1' for the first variation, 'U2' for the second, and so on.
What is the process for including an image in a prompt?
-To include an image, users can find an image online, right-click to copy the image address, and paste it at the beginning of their prompt. They then need to invoke the image within the prompt, for example, by writing 'man in style of Terminator'.
How can users improve the accuracy of their prompts?
-Users can improve prompt accuracy by using a template that clearly defines the desired output, which helps Mid-Journey understand exactly what is needed instead of just receiving raw text inputs.
What is the recommended approach for creating two separate characters in Mid-Journey?
-Creating two separate characters can be challenging as Mid-Journey often blends them together. The recommended approach is to re-roll the prompt multiple times to achieve the desired result.
How can users create photorealistic images with Mid-Journey?
-To create photorealistic images, users should include details in the prompt that indicate a high level of intricacy, such as 'Ultra HD photography' or 'cinematic lighting'.
What is the 'candid photo' prompt style and how does it affect the image?
-The 'candid photo' style means the subject is not looking directly back at the viewer, which is different from the default style of Mid-Journey generations. It creates a more natural and spontaneous feel to the image.
How can users create anime-style images with Mid-Journey?
-To create anime-style images, users can add 'dash dash niji' at the end of their prompt. This transforms the image into an anime style, which is particularly effective for action poses.
How can users change the aspect ratio of their images?
-Users can change the aspect ratio by adding 'dash dash AR' followed by the desired ratio at the end of their prompt, such as 'dash dash AR 3 by 2' for a 3:2 aspect ratio.
What is the benefit of playing around with aspect ratios in Mid-Journey prompts?
-Experimenting with different aspect ratios can result in entirely different images. For example, a 3:2 ratio may favor action poses, while a 2:3 ratio is more like portrait mode, offering a variety of visual outcomes.
Outlines
🚀 Introduction to Mid-Journey: AI Image Creation
This paragraph introduces the concept of Mid-Journey, an AI tool that can create new images based on user input. It explains that users can prompt the AI with text descriptions, combine text with images, or mash two images together to generate unique outputs. The speaker invites viewers to subscribe and like the video, and then provides a step-by-step guide on how to use Mid-Journey through its Discord channel. It also mentions the possibility of adjusting prompts to refine results and incorporating images by using specific commands and formatting. Additionally, the speaker promotes Pirate Tea, a natural energy drink, as a sponsor of the video.
🎨 Tips for Effective Use of Mid-Journey
In this paragraph, the speaker shares six tips to effectively use Mid-Journey. The first tip suggests using a template for more precise control over the image generation. The second addresses the challenge of creating distinct characters by recommending frequent re-rolling. Third, the speaker advises including detailed descriptors for photorealistic images. Fourth, incorporating camera perspectives can enhance the realism of the generated images. Fifth, the use of 'niji' mode is recommended for anime-style images and action poses. Lastly, adjusting the aspect ratio can lead to different outcomes, favoring either action poses or portrait styles depending on the ratio. The speaker concludes by expressing their personal enthusiasm for Mid-Journey and encourages viewers to explore other related content.
Mindmap
Keywords
💡AI
💡Text Prompts
💡Image Mashing
💡Discord
💡Photorealistic
💡Perspective
💡Anime
💡Aspect Ratio
💡Energy Drink
💡Sponsorship
Highlights
Mid-journey is an AI that creates new images based on user input.
Users can prompt Mid-journey with text descriptions like 'Samurai painting in the style of Edward Monet'.
It's possible to combine text and images for more specific prompts.
Mashing two images together is a feature that can create unique combinations like a pet merged with a famous monument.
Mid-journey is available for free and can be accessed by signing into their website with a Discord account.
To create an image, users can type '!imagine' followed by their prompt in the Discord channel.
Each prompt generates four variations, and users can select the one they like by upscaling.
Images can be included in prompts by copying and pasting the image address at the beginning of the prompt.
For prompts with two images, the main image link should be placed first, followed by the second image link.
A template can be used for more accurate control over the prompt without blending elements together.
Creating two separate characters can be challenging, and re-rolling is recommended if they blend together.
Photorealistic images can be generated by including high detail and specific style prompts such as 'Ultra HD photography'.
Camera perspective in the prompt can affect the output, like 'GoPro perspective' or 'drone perspective'.
Using 'candid photo' in the prompt will result in subjects not looking directly at the viewer.
For anime-style images, add '--niji' at the end of the prompt.
Changing the aspect ratio of images with '---AR' followed by dimensions can alter the style of the output.
Different aspect ratios can lead to entirely different outputs, favoring action poses or portrait mode.