Shulou(Shulou.com)11/24 Report--

Thanks to CTOnews.com netizen Sancu for the clue delivery!

The newly upgraded Midjourney makes the whole network crazy again, the creativity is lifted, and the picture is more realistic.

A month ago, a perfect Chinese couple painted by Midjourney V5 went viral online, causing many people to exclaim that the painter is going to lose his job.

Just yesterday, Midjourney announced that V5 is free of charge for a limited time, and it is the latest version of V5.1.

Although it is updated on the V5 version, the upgraded V5.1 creativity is released, and the picture is more silky, which is simply a visual tipping point impact.

Not only can you use colloquial prompt, but also open a new mode of RAW.

Needless to say, the editor immediately tried the "perfect couple in China" and used the latest style raw.

The painting style is amazing in an instant, and you can see that the latest V5.1 is more accurate in dealing with the details of the characters and the colors are more vivid.

After using it, netizens said, "the details of the characters in V5.1 have gone crazy to the extreme." It's perfectly possible to pretend to be real. "

Is it true that human painters really exist?

It's even more amazing, everyone OMG people! If you look at the feelings of netizens, you can see how much this thing is worth a try.

Fewer prompts, more amazing paintings.

5 prompts for "Van Gogh style fashion photography"

Some netizens gave the following cue on Midjourney: "Van Gogh Wind" fashion photography.

I didn't expect that the four pictures generated were so amazing.

Netizens said that if this picture is thrown into the designer community like Behance, it will go viral. However, he also believes that AI is so strong that it will stifle human creativity.

Give three more words a broken hallelujah, plus-- ar 4:3:

Similarly, there are foreign netizens to Van Gogh, Monet and other artists into the style of fashion photography, the effect is very wow, instantly feel that photographers no longer exist.

Van Gogh wind

Basquiat graffiti style

Monet wind

In addition, Midjourney V5.1's "style raw" is really excellent in film quality. The following different images recombine prompt.

A "Cleopatra" prompt:

"film noir mystery film still, low-angle medium closeup two-shot of 2 ancient Egyptian women opening a large gold treasure chest, kohl-lined eyes, bold lip colors, elaborate wigs, and flowing garments, ochre & gold tones, atmospheric lighting, dramatic light and shadow contrast, hieroglyphics, clues, hidden treasure, suspenseful-- ar 16:9-- style raw"

And the control of science fiction style, it really smells like an American blockbuster.

"2000s sci-fi film still, medium full shot, two-shot of 2 astronauts opening a strange artifact, blue & purple tones, atmospheric lighting, dramatic light and shadow contrast, screens with spaceship ui/ux, clues, discover, suspenseful-- ar 16:9-- style raw"

Prompt: "A woman in a red dress is facing you, Underwater photo, hyper photo realistic, enchanted environment, cinematic, 4k, Epic,-- ar 16:9-- v 5.1-- v 5.1"

Some netizens said that V5.1 is more detailed, but more 2D.

The editor of this picture suggests that readers search the original picture on the Internet.

All the details, design, and overall artistic conception of the whole picture are exquisite.

The goddess of Chinese style.

Others found that Midjourney V5.1 seems to understand "bright colors" better than previous versions.

This is the color contrast between V5 and V5.1, you can see that the overall brightness is higher, a bit of a HDR feeling.

Some people continue to focus on their hands, which are even more beautiful.

This, no more description, blockbuster level.

Giraffes!

Let's take a look at the following blockbusters of waste Capital Wind to see if there is any feeling of heartbeat.

V5.1, an epic update, in fact, on May 4, Midjourney quietly launched the new version of V5.1.

If you want to use V5.1, you can set it directly in / settings:

According to officials, V5.1 is easier to use than the previous V5, with these changes compared to V5:

-higher consistency

-more accurate for text prompts

-fewer unwanted borders or text artifacts

-higher clarity

In addition, V5.1 supports simpler prompt, produces more creative graphics, and has recently launched style raw mode.

The function of this raw mode is to maintain the logic of V5, because V5.1 is more creative when creating pictures, and will judge the content and form of the image more according to its point of view.

Sometimes, the images generated by V5.1 sometimes do not meet your expectations.

Just turn on style raw so that the background AI completion function is disabled.

Netizens generated a series of pictures comparing V5.1 with the previous version: (left V5, right V5.1)

It can be seen that the brightness looks more uniform and natural. In particular, look at the backlight between the model's head and the rabbit, and the difference between hair and hair.

Color saturation, the difference in food does not come, the right side is simply mouth-watering.

Netizens showed the RAW model differences of Midjourney v4, v5, V5.1, V5.1.

Midjourney said that in the next few days, data will be collected and analyzed to improve the existing system.

The goal is to convert V5.1 and AI tuning to the default mode by the end of the week, followed by V5.1 style, as well as potential improvements to V5.2 style.

Comparing the upcoming 3D version with the previous version, we can see that V5.1 is completely better than the previous versions in detail, fidelity and color.

But what is even more exciting is to generate video.

There are already some drawing tools that netizens can't wait for text to video.

Come to think of it, as long as you enter text, you can produce a video blockbuster as shocking as the pictures above. What an experience that is.

Coincidentally, there is news that Midjourney is doing this.

In the future, it will take a second to generate a blockbuster that is lifelike to kneeling.

In fact, some companies are already doing so.

In March this year, Runway released a text-generated video model Gen-2.

With Runway Gen-2, you can create a cool blockbuster with any image, video or text, and you can have whatever style you want.

The predecessor of Gen-2 is Gen-1, the first AI editing model launched in February this year.

At that time, Gen-1 can edit the video we want on the basis of the original video. Whether it's a rough 3D animation or a rickety video shot on a mobile phone, Gen-1 can be upgraded to an incredible effect.

However, the visual effect of Gen-1,Gen-2 may be a little worse than that of Gen-1 when editing a video. After all, this is making a video out of nothing.

I believe that the future Midjourney version of Vincent video tools will be another big bomb.

Reference:

Https://bootcamp.uxdesign.cc/midjourney-v5-1-is-out-and-taking-ai-art-to-the-next-level-a5b63026c4

Https://twitter.com/search?q=Midjourney%20V5.1%20&src=typed_query

Https://simonwillison.net/2023/May/4/midjourney-51/

Https://twitter.com/midjourney/status/1654651857552408576

This article comes from the official account of Wechat: Xin Zhiyuan (ID:AI_era)

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.