① Summary

I think learning picture tips is similar to learning drawing, and the best way to learn is not to use templates directly.

Instead, copy the real map or the map generated by others. If English is not good, you can also write Chinese first, and then let ChatGPT translate it. When you copy several copies, you will gradually understand how to make similar drawings.

② case

Take a simple "handshake diagram" as an example. Let's take a closer look at this chart. What are the elements in the diagram?

The first theme is two hands, and then they hold together and look like two Asian men.

Second, both of them are wearing suits.

The third background looks like the door of an office building, and the two may be shaking hands and saying goodbye. And the background is deliberately blurred, or taken with a camera.

To sum up, general information:

Theme: Two Asian businessmen in suits and ties are shaking hands and saying goodbye.

Scene: At the entrance of the office building

Image style: stockphoto camera shooting

At this time, we can try to write tips. If you think your English level is not good enough, you can also try to use translation software: stock photos of two Asian men shaking hands in suits and ties, saying goodbye in front of the main office building, taken with Canon.

Midjourney generates the following results. Emm？ It seems that it is not what we expected. Don't panic, you will encounter this problem when you first start using Midjourney. It's important to try more.

Let's analyze why Midjourney produces such a picture.

According to recovery:

1. First of all, we only need to "shake hands" with the main body or focus of the picture, not two people.

2. The camera mode can't seem to blur the background? More like an image style, Figure 4 is like an old photo style.

Then let's adjust the tips, increase the focus and blur the background keywords:

Pictures of two Asian men in suits shaking.

Hand, say goodbye in front of the main door

Office building. Focus on both hands

Canon, background shot

The results generated are much better. Graphics 1 and 4 basically meet the requirements, and I can also meet our requirements by cutting out graphics 2 and 3. However, it should be noted that there are still some problems of manpower generation in the middle. If you look at Figure 2 and Figure 4 carefully, one of them has six fingers, but I think it should be repaired in the future.

3 Summary, then let's summarize the tips and divide them into the following parts:

The first part (red line): describe the content you want.

Part II (Blue Line): Describe the background/environment of the subject.

The third part (yellow line): the focus position of the photo.

Part IV (Green Line): Original film style or special requirements.

Now, have you summed up a template?