HOME | DD

Description

It’s been a few months since I have had time to post anything (I have been rather busy with an assortment of other projects). Since this time Midjourney has released both V6 Alpha and V6 Niji and while I frequently play around with these developments I haven’t had the time to showcase anything from my exploration until now.

To begin Midjourney V6 is quite a bit of an improvement over there prior V5 and V5.2 releases in terms of its contextualization abilities or in other words if prompted to generate a dog on a fence next to a sign with the text “Meow”, it now understands how to generate the prompt better.

While I do enjoy some of the V5 and V5.2 ascetics over V6 in niche places, V6’s ability to contextually understand how text should be assembled and presented goes beyond anything other Midjourney models have offered before and such focus comes as no surprise given the AI communities overall migration towards context and picture quality as we have seen with releases of Stable Diffusion XL and Dall-E 3.

Keeping this in mind, I decided to showcase some of the new V6 features by once again focusing on a familiar subject to those that follow my posts (the folks schnauzer) and decided that the dog was feeling a little trollish today (Meow it is). Since Kodachrome photography has been a passing interest to me lately I decided to stylize towards this look and feel and after a few iterations in Midjourney I ultimately ended up with something that I liked and, after some minor editing in Photoshop, I present these results here for fun... Enjoy!

Description

It’s been a few months since I have had time to post anything (I have been rather busy with an assortment of other projects). Since this time Midjourney has released both V6 Alpha and V6 Niji and while I frequently play around with these developments I haven’t had the time to showcase anything from my exploration until now.

To begin Midjourney V6 is quite a bit of an improvement over there prior V5 and V5.2 releases in terms of its contextualization abilities or in other words if prompted to generate a dog on a fence next to a sign with the text “Meow”, it now understands how to generate the prompt better.

While I do enjoy some of the V5 and V5.2 ascetics over V6 in niche places, V6’s ability to contextually understand how text should be assembled and presented goes beyond anything other Midjourney models have offered before and such focus comes as no surprise given the AI communities overall migration towards context and picture quality as we have seen with releases of Stable Diffusion XL and Dall-E 3.

Keeping this in mind, I decided to showcase some of the new V6 features by once again focusing on a familiar subject to those that follow my posts (the folks schnauzer) and decided that the dog was feeling a little trollish today (Meow it is). Since Kodachrome photography has been a passing interest to me lately I decided to stylize towards this look and feel and after a few iterations in Midjourney I ultimately ended up with something that I liked and, after some minor editing in Photoshop, I present these results here for fun... Enjoy!