The Midjourney v5 language model adds realism to human hands

Midjourney v5 is the latest language model of the popular text-to-image generator known for its realistic creations.

The update rolled out to Midjourney’s paid customer base on Wednesday, and many users, including graphic designer Julie Wieland, have shared their new AI-generated artwork. AI details brought by the v5 language model include improved “efficiency, coherence and quality,” Midjourney said on its website.

Just a warning – Midjourney’s AI can now make hands right. Be especially critical of political images (particularly photography) you see online that are trying to provoke a reaction.

โ€” Del Walker (@TheCartelDel) March 16, 2023

Many users have particularly noticed the improvements in the detail of the human hand, with subjects closer to having five fingers per hand. In previous models, human hands were often generated with an incorrect number of digits, anywhere from four to between seven and ten. Those who were able to test Midjourney v5 were impressed with the visual upgrades.

“The latest Midjourney v5 model is both extremely mind-blowing/scary and beyond fascinating,” Wieland told PetaPixel.

“His ability to recreate intricate details and textures like realistic skin textures/facial features and lighting is unparalleled,” she added.

While Midjourney is a somewhat unique language model, it works in a similar way to other text-to-image generators like Stable Diffusion and DALL-E, using formulated prompt descriptors to create images according to the model, which is trained with humans – made art for reference, according to ArsTechnica.

There has been an ongoing debate about whether AI-generated art is legal or ethical; However, many Midjourney v5 users feel that as upgrades continue, it may become indistinguishable from real art and photography. Wieland believes that AI image generators like Midjourney should be companions rather than substitutes for photographers.

READ :  ARway Corp. commissions Stockhouse, Canada's leading finance portal

Midjourney v5 is here! (really this time, lol)

Here are some of my prompts side by side, v4 vs v5, plus some new prompts and bulk shots. I’ll add more as I experiment.


โ€” Nick St Pierre (@nickfloats) March 15, 2023

However, Midjourney v5 is not without its flaws. AI image creator Nick St. Pierre has shared much of his work using the same prompts in the previous Midjourney v4 and Midjourney v5. One image was an intriguing shot of young women in 1960s street style, reminiscent of an outdoor fashion shoot. A closer look reveals the floating hands, not attached to bodies, and misshapen faces in the background.

With his latest v5 model, Midjourney founder David Holz recommended on the brand’s Discord that users should shy away from short prompts in favor of “longer, more explicit text” to get better results.

Editor’s Recommendations