Creating an anime profile picture is as easy as using these three AI apps

The advent of AI-based tools gives us a thousand and one ways to generate original content. In this sense, some solutions such as Stable Diffusion or Lensa stand out, but if there is one that has revolutionized this fieldit’s Midjourney.

When we create images with these applications, I thought more than once to test them, and there is no better way than to request a character for an anime. As a fan of Japanese anime series, I tell you How to do it and I show you several results that left me speechless.

Create the next Tanjiro or Monkey D. Luffy

In the minds of many anime fans, there has always been the idea of ​​imagining new characters who could join our favorite fictions. Let’s face it, we’re real fans (some would call us weird) but nothing could be further from the truth, now it’s possible thanks to robots What let our imagination fly.

I tried three of the most popular tools: Midjourney, Stable Diffusion and Dall-E2. And the result could not be more disparate. I make a spoiler and that’s it Midjourney continues to post the best results

, also with difference. Obviously, each application uses different methods to generate images from text, so it’s only natural that they provide different results.

It should be mentioned that the fast introduced was the same in all three utilities, so do not disadvantage any of them. I used an English query because these generators understand Shakespeare’s language better. More specifically, the text introduced was: A blonde-haired, green-eyed anime character in the style of Attack on Titan.

We are starting to look at results with Stable Diffusion, the Stability AI solution that is open-source. As we can see, the images are of quite acceptable quality, with some detail. However, not everything went as it should with a somewhat strange character (first image from the left side). Almost any of these characters could be the next protagonist in an anime or manga, and the truth is, they’re pretty cool.

Now let’s move on to DALL·E 2, developed by OpenAI, the same people who created the famous ChatGPT and now has a paid version. This tool returns a single image. Without fear of being wrong, we can affirm that it is the worst result by far. While it’s not a bad design (I couldn’t create such an image), it has a somewhat basic design for what these utilities are used to.

To end this round, we see what Midjourney is capable of. As we predicted earlier, this is the one that best resolves the image. Or rather the images, since Midjourney offers a set of up to four elaborations. As we can see from the image below, some of them seem to be taken from a real anime. Each of the four results is mouth-watering and has incredible design and level of detail. Midjourney demonstrates the impressive capacity of its Artificial Intelligence.

With a little imagination you can become an anime character

As we have seen, advances in AI not only allow us to have chatbot like Bing with ChatGPT, but create images for any purpose. For example, if we are looking for a profile picture, a wallpaper or simply a design for a personal project, these applications make our life easier.

The most interesting thing about it is that we don’t need a computer, we can use them from mobile

, no problem. Midjourney can be accessed via Discord and the other two apps have dedicated websites, which we can access from our phone’s browser. To enter the fast obviously we will use the keyboard, and the generated images can be perfectly saved in the internal memory.

If in your case you want transform into anime character, You can do it too. To do this, you will have to upload your own photo beforehand with the request, yes, do not forget to tell the AI ​​that you want an anime style image (or with any other condition you wish). In this case, Stable Diffusion is an exception and does not allow uploading images, so we forgot to use it for this conversion.

In DALL-E, its main screen contains a button for uploading images. You have to touch Upload an image then select the photo you want to convert. Then enter the fast and knock Generateafter a few seconds your image will be ready.

In Midjourney we can use this same functionality. To do this, in Discord, press the “+” button to upload a photo, then type /imagine and followed by, enter the link of your photo, then the request. When ready, the robot it will turn you into a very cool anime character.

As we have seen, we can create real wonders thanks to new trends in Artificial Intelligence. Having our own character or transforming an anime-style profile picture has never been possible, until now.

In Xataka Android | The fastest way to get ChatGPT on your mobile: Bing makes it easy for you to get it