AI-Creator ai can generate any image with only text prompts. Such a tool allows everyone to enter any description they want freely, so as to obtain funny, amazing pictures
AI-designer , can modify photos with prompts, like changing background or replacing a part in pics, which can be a good assistance to those who have a high requirement for photos
Enter the official website link(https://imgcreator.zmo.ai ), and fill a text description in the blank. For example：a little boy is flying a kite. You can even enter a poem or lyrics, no matter what you can imagine, AI will generate it for you
As a new comer, you may worried that you are unable to use prompt (keywords) accurately，then the results will be disappointing. However,the biggest difference between imgcreator and other AI drawing tools is that there is no need to use specific magic words, just describe in simple keywords or sentences. The following are the skills I have summarized
1. Scene description = [adjective/adverb + subject] + background description + details You need to describe as clearly as possible what you want to present: the background, the details. Different specific descriptions and styles will directly determine the effect of the picture. Here are some examples:
If you have no inspiration for the time being, you can also click "Try an example" at the top right of the texting box to randomly generate keyword sentences for users to learn. Just start copying directly!
The artworks that imgcreator generates concludes six categories:Anime, Avtor,Art, Realistic photo,Design,Illustration. Selection of category will directly affect the final quality and style of the picture. Pictures with the same description but different categories have completely different visual effects on users.
Each category has more than four styles for users to select, you can choose your favorite style according to the scene you describe and the category you want the picture to present. Including the classic style like Ukiyo-e, Van Gogh, Monet and so on. What's more, it has Hayao Miyazaki's style and One Piece style! Just entering the corresponding keywords, you don't need any art foundation at all.
Combining the above three techniques, I believe you have mastered the basic function of this tool. But there are a few tips to note:
- Due to the different algorithm models, if you blindly pile up the "mantras" that are commonly used in other products, such as octane rendering, plasma, it may be counterproductive
- At the same time, your description also needs to be consistent with the category and style you choose. For example, if you choose the style of "Van Gogh", while your aim is to generate a realistic photo, the final result may not be satisfying
- If you want to enrich the details of the characters, you can add "close-up"
- Too complex, lengthy descriptions may generate incorrect pictures and texts. Example: 'Queen Medusa,Muscular man, Skeleton, the chains were entangled with wild wolves,countless insects with dragonfly wings like purple centipedes floated in the air. From the cracks in the rock, there were huge colorful snakes
Observed from the result, it seems that AI has a long way to go to understand complicated sentences
I think you might have already noticed that there is a 'Freeform'. Freeform format doesn't need users choose category or style. You can just type what you want. But please make sure you have input a highly detailed prompt, otherwise you will be unsatisfied with the result
For example :3d octane render, stunning beautiful sweden adult female with profile face, rim light, big leaves and stems, gold roots, gold fine foliage lace, alexander mcqueen, gold art nouveau fashion embroidered collar, steampunk, silver
Equipped with PS function is one of our highlights. Sometimes, the results generated by the AI may not meet your requirements, or there are places where you want to modify in order to achieve a more perfect result. At this time, the writer strongly recommends that you take full advantage of imgeditor ( https://imgeditor.zmo.ai ) to improve.
With only three steps, you can become a master of Photoshop.
As we can see from the result, AI not only helps me replace that painting, but also out-painting my original pics. Space in the room is larger. Surprisingly, it finished within a few seconds
However, nothing is perfect. After testing countless pictures of neither fish nor fowl, I finally got the following secret recipe!
If you want to edit one part of a close-up picture, please make sure that part occupies enough space. Then AI will assist you better to identify
If the background of original image is white, you need to use the [Erase] tool to erase the white background and enter the prompt to generate, otherwise you will get the original image
If people's faces are small in your image, changing background may be a better option.The writer recommends that you do not randomly try to change the face or the hands of the figure, because the body parts of real humans seem to be the knowledge blind spots of AI
Currently, three sizes of images are supported, which are similar to 1:1, 16:9 and 9:16.
- The first size is recommended if you want to generate images of people or objects with outstanding detail
- If you want to turn the generated image into a video, it is recommended to use the second and third sizes, which are compatible with short videos and long videos respectively
- Of course, you can choose the size you want, we won't limit it in case there are surprises
At present, we provide 1k images by default, and of course we will provide users with 4k pixel options in the future~
In your history, you can delete that photo you are unsatisfied with
The development path of AIGC technology still needs to be explored by experts. Of course, as a user, you are also indispensable to witness. Bring your imagination and become the best magician in "Hogwarts"