Midjourney Introduces An Image Editor And A Re-Texturing Feature For Better Creative Customization

The AI field, once relatively dull, has been completely disrupted.

The launch of ChatGPT by OpenAI sent shockwaves through the industry, sparking both excitement and concern. It triggered an arms race among tech companies, big and small, all vying to develop the most advanced Large Language Models and improve them at breakneck speed.

What was once a mundane topic has now transformed into a booming, highly lucrative industry.

Midjourney is one the smaller players.

But despite its relative smaller influence if compared to larger rivals, Midjourney has generated many of the most realistic AI-made imagery ever created.

And this time, things just get better.

Read: How Midjourney AI Art Generator Creates Photorealistic Images, Without The Nightmarish Hands

Midjourney is introducing two new features to its AI image generator: an image editor for uploaded images and an image restructuring tool.

While the former describes what it does, the latter is more about giving users the ability to restructure, or modify elements of an image they've generated.

For example, users can change the looks of the surface, modify elements such as the materials, change lighting, and more.

Both features integrate seamlessly with Midjourney’s existing customization options, including style references, character references, and personalized models.

However, users may encounter some early-stage issues, such as unintended changes in small areas or disproportionate figures with small heads.

To ensure a smooth rollout, Midjourney is allowing its community, human moderators, advanced AI moderation systems, and server infrastructure to gradually adjust to these new capabilities.

Initially, access is limited to annual subscribers, users who have joined within the past 12 months, and those who have generated at least 10,000 images.

These features are made possible thanks to Midjourney's breakthrough achievement, where its AI can create consistent characters across multiple AI-generated images.