drawing skills
Generally speaking, those who engage in technology and write articles often add zero points to the skill tree of painting.
Moreover, skills such as painting and music often require a certain amount of talent. It is not like writing codes and excavators. The latter two can often be learned by everyone after training. However, if the former two do not have a little talent, it is often difficult after training. learn. And in real life, everyone is very busy with work and study, so how can there be so much time to add points to so many skill trees.
The emergence of AI tools such as ChatGPT and Midjourney allows us to add skill points in areas that we are not good at in just a few minutes, turning ourselves into a hexagonal warrior.
The rapid development of AI has made many people feel anxious. In fact, it is often not AI that will replace us in the future, but other people who master AI tools. So, from now on, follow up and learn AI tools quickly, and you will be more competitive than others.
AI painting example
Let's enjoy some pictures first, these are drawn with Midjourney:
It can be seen that Midjourney can not only draw pictures similar to the real shots of the camera lens, but also draw pictures similar to those produced by CG special effects, and can also draw pictures in the style of hand-painted illustrations. , oil painting, gouache painting, abstract painting, sketch, etc., Midjourney can draw them all. Due to the limited space, I will not give examples one by one.
illustration practice
Next, let's practice. Our most urgent need at present is the illustrations in the article.
In the past 10 years, I often wrote a lot of articles and tutorials on blogs and Zhihu, but there were often only dry texts, especially when the text length reached more than a thousand words. One has the patience to read it carefully. At this time, the accompanying pictures in the article are particularly important.
But with pictures, you can't match them blindly. If you match some landscape pictures, it looks good, but it has nothing to do with the text content, so it won't work. In addition, when searching for pictures on Baidu, there are often two problems. One is that the pictures are difficult to match the text content, and the other is that it is easy to duplicate and infringe on others. And I can't draw by myself. Let me say that I don't have the ability to draw. Even if I do, it usually takes a few hours to draw an illustration. Wouldn't it take a few days to write an article? Difficult.
cue word construction
With AI drawing tools, such as Midjourney, it is much easier for us to match pictures when writing articles. If we write an article about the skills of taking notes, then we need a picture with a pen and a notebook on the picture, then we construct the prompt words like this:
The first is the painting style description, we can use these words:
flat design (flat design) cartoon (cartoon) illustration (illustration)
Then the screen content description:
pen (pen) notebook (notebook)
Then we combine the prompt words:
flat design, cartoon, illustration, pen, notebook --ar 3:2
Where –ar 3:2 means the aspect ratio of the generated image, here we hope to generate a horizontal version with a ratio of 3:2
Then Midjourney gave us the result based on these prompt words:
Of course, you can continue to let the AI redraw, and it will give different results each time:
If we want the screen elements to be simpler, we can remove the cartoon elements and use prompt words:
flat design (flat design) simple (simple) illustration (illustration)
Then we combine the prompt words:
flat design, simple, illustration, pen, notebook --ar 3:2
It can also be seen from the picture that the cartoon elements have been moved, which is more suitable for serious content articles
If we want two cats to be drawn on the notebook, we only need to add the corresponding prompt words:
flat design, cartoon, illustration, tow cats, pen, notebook --ar 3:2
Then, if we write an article about digital cameras, we can construct prompt words like this:
flat design, cartoon, illustration, digital camera --ar 3:2
If you're writing an article about your cat being playful, you can structure your prompts like this:
flat design, cartoon, illustration, cat, play ball --ar 3:2
You can see that AI draws a picture of a cat playing with a ball according to our requirements.
Of course, if you don't like this cartoon illustration style, you can adjust the prompt words to guide AI to generate pictures of other styles.
More styles of AI painting, we will introduce them one by one in the next article.
Embrace AI
We don't know what occupations AI will replace in the future, but we are certainly not the first to be eliminated if we master the use of AI tools.
By the way, I also used Midjourney to generate the picture on the cover of this issue.
Welcome to pay attention to the WeChat public account [Kicked Lao Li], we will continue to share the content of AI tools to improve work efficiency and enrich personal skills.