The feature is said to enable users to modify parts or details of a generated image without having to recreate an entirely new work. PCMag reports that, for example, users can use the tool to create an image of “a fish jumping out of the water,” and then use the “Vary (Region)” button to enter a new prompt, such as replacing the previously created fish with a shark. Replace the previously generated fish with a shark, for example.

Midjourney says this feature works best when the selected region is 20-50% of the image size. Additionally, it works better if the local details of the change match the image better.

For example, as shown in the image below, Chase Lean, one of the users who participated in the beta test, claims that the feature can change a model’s clothes, and demonstrates the effect of changing her clothes from a blue sweatshirt to a black sweatshirt to a variety of fashion/professional outfits, with little or no change in the background or the model’s pose.

Wide

The post AI painting tool Midjourney introduces an Inpainting feature appeared first on TechGoing.

▲ Tweets posted by the source Midjourney

It is reported that the Midjourney 5.2 version brings three different styles of zoom simulation for the “Zoom out” function, providing three zoom options, which are 0.5, 1.5 times, and 2 times the original image “lens focal length”. That is, if the original image (1.0x zoom) uses a focal length of 50mm, it can generate perspective effects between 28-35mm, 75mm, and 100mm.

▲ Picture source OrctonAI

▲ Picture source Midjourney

It is noticed that Midjourney version 5.2 also provides other functions, such as “Make Square”, which can compress the length and width of the picture in a perspective style to make it a square.

▲ Picture source Midjourney

The new function “Variation Mode” allows users to fine-tune the visual difference of the image, as shown in the figure below.

▲ Picture source Midjourney

In addition, the user can also use the stylize parameter to simply modify the style of the generated picture, and analyze the prompt words to let the user know which prompt words will not have any effect on the output image, and which prompt words can dominate the output image effect.

The post AI painting tool Midjourney welcomes version 5.2 update, can generate camera zoom effect appeared first on TechGoing.

This feature allows users to upload any image, and MidJourney will automatically analyze the content of the image to generate 4 suitable prompts and 4 buttons, which can be clicked to generate more similar images based on these prompts, or to add their own ideas to these prompts to create more interesting works. This way users don’t have to spend too much time conceiving and modifying Prompts and can get satisfactory drawing results faster and easier.

The subscribers can experience this feature by using the /describe command in MidJourney’s Discord server, where typing /describe on Discord will prompt you to upload an image. This feature is so powerful that some users have used it to redraw some of the world’s most famous paintings.

The post AI drawing tool MidJourney update: can generate Prompt based pictures appeared first on TechGoing.