AI繪圖一次就上手!深入實作教學 Midjourney niji journey ChatGPT v5

蘋果妹
28 Apr 202334:22

TLDRThis video script introduces Midjourney, an AI drawing robot that transforms text prompts into images, and its complementary use with ChatGPT for text processing. It covers the basics of Midjourney, its main features, and various commercial applications like graphic design and game art. The script also explains the process of generating images using Midjourney, including setting up the AI, inputting prompts, and utilizing different modes for various styles. Additionally, it discusses the integration of ChatGPT to enhance the prompt generation process, the importance of settings like aspect ratio and style value, and the potential for private creation spaces and alternative software for a free trial. The video concludes with a discussion on pricing plans and the potential for future updates and tools.

Takeaways

  • 🤖 Midjourney is an AI drawing robot that processes text prompts into images, complementing text-based AI like ChatGPT.
  • 🎨 It has a wide range of commercial applications including graphic design, game art, film concept art, and realistic photography.
  • 🌐 Midjourney operates through Discord and can be used by anyone with a web browser on mobile or computer.
  • 🚫 Midjourney's free trial is currently suspended due to abuse, and the service offers paid plans for continued use.
  • 🎮 Paid plans include options for monthly usage limits, unlimited generation, and private workspaces.
  • 🖌️ The creative process involves entering prompts that include themes, styles, mediums, and atmospheric descriptions.
  • 📸 Midjourney offers different AI models (V4 and V5) that excel in different styles, such as realistic or atmospheric images.
  • 🔄 The AI can generate variations of an image by adjusting the 'Style value,' controlling the divergence from the original prompt.
  • 🤹‍♂️ ChatGPT can be used to generate high-quality prompts for Midjourney, speeding up the image creation process.
  • 🌍 Alternative AI tools like niji journey and Bing Image Creator offer similar functionalities with different focuses and styles.
  • 📦 Midjourney's official gallery allows users to view and compare their works with others, providing a community aspect to the platform.

Q & A

  • What is the primary function of Midjourney?

    -Midjourney is an AI drawing robot that primarily functions to convert text prompts into images, processing input text into visual content rather than generating meaningful text content like ChatGPT.

  • How can Midjourney be used in commercial design?

    -Midjourney can be utilized in various commercial applications such as graphic design, game art design, film or animation concept art, and realistic photography. It can assist in creating drafts for logos, website designs, and packaging designs.

  • What are the main features of Midjourney?

    -The main features of Midjourney include its ability to process text commands, generate images based on English words or phrases, and create a wide range of visual content from game art to movie-style pictures.

  • How does Midjourney differ from ChatGPT?

    -While both are AI systems, Midjourney focuses on image processing and generating visual content from text prompts, whereas ChatGPT is designed for text processing and can generate meaningful text content.

  • What is the significance of the 'Prompt' field in Midjourney?

    -The 'Prompt' field is where users enter the keywords or text descriptions that Midjourney uses to generate images. It is essentially the input field for telling the AI what kind of picture to create.

  • How can users enhance their Midjourney experience with ChatGPT?

    -Users can leverage ChatGPT to generate high-quality prompts for Midjourney. By utilizing ChatGPT's ability to understand and generate text, users can receive prompt suggestions that can then be input into Midjourney to create images.

  • What are the different 'creation modes' available in Midjourney?

    -Midjourney offers various creation modes, including MJ version 4 and MJ version 5 (V4 and V5), which use different AI models to generate images. V5 is known for producing super-realistic photos, while V4 is better for creating a sense of atmosphere and personality.

  • How does the 'Style value' setting in Midjourney affect the generated images?

    -The 'Style value' controls the degree of divergence of AI creation. A lower Style value generates more conservative images according to the prompt, while a higher Style value allows for more creative freedom and potentially more varied styles in the output.

  • What is the 'niji journey' and how does it differ from Midjourney?

    -Niji journey is a twin of Midjourney that specializes in generating Japanese anime-style pictures. It accepts Chinese input and produces images in the anime style, catering to users who prefer this specific visual aesthetic.

  • What are the different pricing plans for Midjourney?

    -Midjourney offers three main pricing plans: a $10/month plan for about 200 pictures, a $30/month plan for unlimited picture generation with a 15-hour fast generation time limit, and a $60/month Pro plan with additional features like incognito mode and the ability to run up to 12 tasks simultaneously.

  • How can users create a private creation space on Midjourney?

    -Users can create a private creation space by either creating a server and adding the Midjourney Bot to it or by using a direct message (DM) with the Midjourney Bot, which keeps the generated images private.

Outlines

00:00

🤖 Introduction to AI Drawing with Midjourney

This paragraph introduces the viewer to AI drawing, specifically focusing on the AI robot Midjourney. It emphasizes the complementary nature of Midjourney and ChatGPT, with the former being for image processing and the latter for text processing. The video aims to provide a step-by-step understanding of Midjourney, its main features, and its commercial uses such as graphic design, game art, and animation concept art. It also mentions the limitations of Midjourney in generating text content and suggests using third-party apps for text-related issues.

05:01

🎨 Midjourney's Functionality and Settings

The paragraph delves into the specifics of using Midjourney, including the input fields for text commands, the process of generating images, and the importance of Midjourney settings like creation mode and image ratio. It explains how Midjourney converts text into images and provides examples of prompts that can be used. The paragraph also touches on the community aspect of Midjourney, where users can view and compare their works with others, and mentions the platform's Discord integration.

10:03

🖌️ Advanced Usage and Prompt Engineering

This section discusses the concept of 'Prompt Engineering' and the importance of the order and choice of words in prompts. It explains how the theme, style, medium, and atmospheric elements are crucial in generating images with Midjourney. The paragraph also covers the aspect ratio settings and provides an example of creating a new Disney princess using specific prompts. It highlights the variability in AI-generated images and the options to upscale or vary the generated images.

15:03

🌐 Midjourney's Gallery and Advanced Settings

The paragraph talks about the gallery feature on Midjourney's official website, where users can view their own and others' works. It explains the process of generating images and the actions that can be performed on them, such as upscaling and variation. The section also introduces more advanced settings like creative modes (MJ version 4 and 5), Niji mode for anime-style images, and the Style value that controls the degree of divergence in AI creation. The paragraph emphasizes the influence of operation mode on the final image outcome.

20:04

🤖 Combining ChatGPT and Midjourney

This part of the script discusses the powerful combination of ChatGPT and Midjourney for generating prompts and images. It explains how to use ChatGPT to generate high-quality prompts for Midjourney, including role-playing with ChatGPT to produce prompts that fit specific themes. The paragraph provides examples of using ChatGPT to generate prompts for logos, web design, and realistic photos, and how to refine these prompts based on the initial results from Midjourney.

25:05

📸 Generating Realistic Photos with Midjourney

The paragraph focuses on using Midjourney to generate realistic photos, emphasizing the capabilities of the V5 mode for this purpose. It discusses the process of pre-booking the type of photos desired, such as street photography or movie stills, and the use of ChatGPT for refining prompts. The section also addresses the limitations of V5 mode in generating perfect images, particularly with eyes and hands, and suggests using third-party apps to enhance the resolution of generated images.

30:05

🏞️ Midjourney's Natural Landscapes and Paid Plans

This section covers the generation of natural landscapes using Midjourney, both with and without specifying a location. It compares the results from different models and highlights the realistic quality of V5 mode. The paragraph also introduces Midjourney's paid plans, including the monthly and annual payment options, and the features of each plan such as the number of pictures allowed, Fast and Relax hours, and the incognito function for private works. Additionally, it mentions the creation of a private server on Discord for generating images and the existence of niji journey, a twin AI for anime-style images.

🌐 Alternatives to Midjourney and Final Thoughts

The final paragraph discusses alternatives to Midjourney for those who are hesitant to pay for the service. It mentions third-party websites that offer similar functionalities and the Bing Image Creator as an option. The speaker shares their personal experience with these alternatives and encourages viewers to subscribe to their channel for updates on Midjourney and other AI tools. The video concludes with a reminder to follow the speaker on Instagram for related content and a farewell to the viewers.

Mindmap

Keywords

💡AI drawing

AI drawing refers to the process of using artificial intelligence to create visual art. In the context of the video, it specifically relates to using AI robots like Midjourney to generate images from text prompts. The AI interprets the text descriptions and produces corresponding artwork, which can be used in various fields such as graphic design, game art, and animation.

💡Midjourney

Midjourney is an AI drawing platform that converts text prompts into images. It is capable of generating a wide range of visual content, from graphic designs to realistic photographs. The AI model is known for its ability to produce high-quality images by interpreting textual descriptions, and it offers different modes for various styles of artwork.

💡ChatGPT

ChatGPT is an AI language model designed for text processing. It can generate human-like text based on given prompts, which can be used for various purposes, such as writing code, creating content, or even generating prompts for AI drawing platforms like Midjourney. The video suggests using ChatGPT to craft detailed prompts that can then be input into Midjourney to produce images.

💡Text-to-image

Text-to-image is the process of converting textual descriptions into visual images. AI platforms like Midjourney and ChatGPT use this process to generate images based on textual prompts provided by users. The AI interprets the text and creates images that match the described concepts, themes, or styles.

💡Prompts

In the context of AI drawing, prompts are the textual descriptions or commands that guide the AI to generate specific images. Prompts can include details about the theme, style, medium, and other elements that the AI should consider when creating the artwork. Well-crafted prompts are crucial for achieving desired outcomes in AI-generated images.

💡Aspect Ratio

Aspect ratio refers to the proportional relationship between the width and height of an image. In AI drawing platforms like Midjourney, users can specify the aspect ratio to determine the shape and orientation of the generated images, such as 2:3, which would result in an image that is two parts wide for every three parts tall.

💡Style and Medium

Style and medium in AI drawing refer to the artistic style and the method or material of the artwork that the AI should aim to replicate or be inspired by. Users can specify styles ranging from Impressionism to anime and mediums like hand-painted illustrations or digital graphic design to guide the AI in creating images with a desired aesthetic.

💡Discord

Discord is a communication platform that combines elements of chat services and community forums. In the context of the video, Discord is used as the community platform where Midjourney operates, allowing users to interact with the AI through text commands in dedicated channels. Discord also features bots that can perform various tasks, including AI-generated content creation.

💡Niji Mode

Niji Mode is a specific setting within the Midjourney AI platform that is dedicated to generating anime-style images. It is designed to produce artwork in the style of Japanese animation, which often has distinct visual characteristics such as colorful and expressive characters.

💡Upscale and Variation

In the context of AI-generated images, 'Upscale' refers to the process of enhancing the resolution or quality of an image, while 'Variation' involves creating a modified version of an image with slight differences, such as changes in facial features or accessories. These features in Midjourney allow users to refine and diversify their generated images.

💡Private Space

Private Space in the context of Midjourney refers to a personal or exclusive area within the platform where users can generate and store their images without them being publicly visible in the gallery. This feature is particularly useful for users who want to maintain privacy over their creations, especially for commercial or confidential projects.

Highlights

Introduction to AI drawing and Midjourney, an AI drawing robot.

Comparison of Midjourney and ChatGPT, highlighting their complementary roles in text and image processing.

Exploration of Midjourney's commercial applications in graphic design, game art, film, and animation.

Explanation of Midjourney's inability to generate meaningful text content, unlike ChatGPT.

Integration of ChatGPT and Midjourney to enhance the creative process.

Demonstration of how to use Midjourney to generate images from basic operations.

Discussion on Midjourney's settings, including creation mode and image ratio.

Overview of Midjourney's AI model and the process of converting text into images.

Introduction to the official website of Midjourney and its connection to Discord.

Information on Midjourney's suspension of free trials due to abuse.

Explanation of the process of drawing with Midjourney, including the input field and function menu.

Details on the Prompt field and its significance in generating images with Midjourney.

Discussion on the importance of Prompt Engineering and the impact of text order on final results.

Description of the general directions set by Prompt, including theme, style, medium, and atmosphere.

Example of generating a new Disney princess using Midjourney and the significance of aspect ratio settings.

Explanation of the variation and upscale functions in Midjourney for refining generated images.

Demonstration of the unique results produced by Midjourney with the same prompt, showcasing its non-repetitive nature.

Introduction to the gallery feature on Midjourney's official website for showcasing and comparing works.

Discussion on advanced usage skills and settings in Midjourney for optimizing image generation.