Say Goodbye to Prompt Engineering: Midjourney V6 Guide!

mid Journey has recently released an alpha test for their latest version V6 and has once again leveled the playing field for all AI image generators say goodbye to long hours trying to figure out prom engineering as V6 has an improved natural language understanding and now allows you to describe your scenes with longer promts to get the exact shot that you want and so today we are diving deep into its new features and how you can effectively prompt it to create cinematic shots of your dreams let's go to start the first feature we will talk about is the natural language understanding it is known to be able to now more accurately read and create what we want according to our description and this new model is more sensitive to the descriptive words we use in the prompt and will not read previous junks like award-winning photo realistic 4K 8K Etc so this means that we can start to really be as explicit as possible and type more naturally in state so to really showcase what this means Let's test it out with simple language see what we can get starting with our first test we will use a descriptive example of an Asian girl with red hair in a diner opposite a blond hair Asian boy and as you can see from the example V6 really does get the details a whole lot more accurate to a very descriptive prom with the correct ethnicity hair color clothing details it is a great Improvement compar to the V5 generation which may mess up the details in the scenes such as giving the go the blonde hair instead it also follows the instruction more clearly making sure that they are actually facing each other like what we wanted to further illustrate how well it could better understand our language I would like to share this example of a fruit basket inspired by clarinet's tutorial on the meat Journey Discord Channel and as you can see from our example the shot given by V6 really follows the prom description more accurately making sure the fruits are in the correct orientation the image given by V6 also does show details of the circular window as mentioned this shows that mid journey is also able to effectively arrange them in order which is super impressive and a huge upgrade and Improvement to the accuracy of the images aligned to what we want besides understanding our descriptions way better it is also able to have a brief understanding of words and texts in the images now so so besides humans let's use an example of a photographic product shot to see how much of a product design we can create and in this case let's say I want to create a perfume bottle for an imaginary brand called Bachi and I want the shot to be on top of the water with flowers like those cool product shots I will try with this prompt over here and here you can see the difference V6 is able to not only bring in the text and car the words onto the bottle nicely it is also able to keep the details of what I want showcasing it lying on the surface of the water and it also creates a really nice reflection of the bottle too and this accuracy is a really big game Cher for product designers because now I can create full product shots in just a prompt and these can be generated and iterated within minutes take another example imagine we want to create a mockup bottle design for a product idea that we have in mind by prompting out the details of what we want one such as having a pink background blue cap and a yellow bottle with our imaginary brand name Da Vinci on it all we have to do is just to prompt it and we can now very easily once again create another beautiful mockup shot for another one of our product ideas and by quickly running them through an AI animation to like peer we can easily create a scene from a product advert take a look and product shots are just one of the many examples we can create with these tool what this means is that we can now create more cinematic shots and scenes for our videos as well and the possibilities are truly endless from here another useful feature of V6 accuracy and natural language understanding is also its abilities to creating storyboards and comic scenes but before we start getting into it here's a prompt tip while V6 understands natural language better with still con describe a long sequence of actions such as three comic book scen of a man that went to a supermarket to get a basket of fruits he got into his car drove home and gave the fruits to his wife as there are too many actions demanded inside it will be flattened to a single snapshot so what we can actually do here is to describe it instead for example a comic book has three panels the first panel depicts a man buying a basket of fruits in a supermarket in the second panel the man is driving in his car the third panel shows the man at the front of his house passing to his wife the basket of fruits and I kept the aspect ratio to 31 as I find it to be the most useful for creating the tree strip comic and here we have something similar to a shot that we want however of course for this specific feature at the moment we do have to give it a couple of tries to get what we really want and we won't get the money shot all the time but it is extremely helpful in helping us out with storyboarding and creating scenes for our videos and as V6 continue to progress for its full release I believe it would be able to bring us closer to generating full-blown comic scenes and story boards at ease moving on besides just upscaling your images another cool new feature is the creative up skill too it provides a creative AI up skill which I think is very interesting for us to get different results for example let's take a look at this rad go interestingly by putting her through the creative AI up skill it seemed to have removed her face piercing and gave her a cleaner look almost like she put on additional Foundations at the moment I do like the one before the creative up skill as it looks less edited and more realistic but the up skill is pretty cool too here's a couple more examples for you to take a look and see the difference and last but not least before we end off the video one the features that people have been debating a lot about recently is mid Journey's ability to better interpret Styles and not just any Styles but the Styles we are familiar with let's take a look at a couple of examples and as we can see from all these different examples the interpretations and the generations are really very very close to the styles that we are very familiar with now before we all start debating I don't think what we would really like to do is to copy the artist work style exactly and I do believe there will be copyright rules in the years to come but what I think is really cool here is the added level of understanding and consistency with its Styles this could really bring us a step closer to generating consistent style sets of our own which will be extremely valuable in helping us creating consistent scenes and shots of our very own unique Styles and that would really be a step closer to our own Army of AI assistant for sure and so these are just some of the very exciting updates that we currently have with me Journey V6 do note that it is still in its Alpha test launch so there will definitely be more changes and updates along the way hence I'll definitely be doing up a follow-up video with a in-depth guide as the full version release so do subscribe and stay tuned for that till then I hope you found this video useful and insightful and if you would like to learn more about AI animations and Creations you may check out my other videos right here peace

Channel: dainami ai

Views: 2,620

Keywords: midjourney, v6, prompt engineering, ai animation, ai art, best ai art generator, best ai tool, ai art tutorial, midjourney v6 tutorial, Midjourney ai tutorial, ai filmmaking, ai generators, ai image generator, cinematic ai art, ai animation generator, dall e, chatgpt, pika labs, runway ml, prompting, ai marketing, ai product design, ai design, ai design product

Id: h0y-zelMOsE

Length: 7min 57sec (477 seconds)

Published: Thu Dec 28 2023