How to use the full power of MidJourney V5 new features. MUST SEE for studio photographers!
HTML-код
- Опубликовано: 6 апр 2023
- I'll show you how to use a new features of MidJourney v5 to get PRECISE images that you need for your work. Example: you take a picture with your phone and turn it into a professional studio-like shot with Midjourney using "describe". And much more..
In this video:
00:52 The beauty of "describe" command and how to use it.
05:15 "Describe" for interior designers.
10:15 "Blend" command (amazing results, you'll be surprised!
13:40 "Blend" + "Describe" + "Reference Image" = insane results
19:07 My new project, www.onlineschoolcreator.com
Check Who is Alex Koloskov: www.onlineschoolcreator.com
🎓 📷 ❤️ Subscribe to Our Channel Photigy: 👉👉 bit.ly/2OPhZdA
Photigy is the interactive school of photography, retouching and CGI. The fastest way to become an expert for everyone in product photography.
Subscribe on bit.ly/Photigy-Website and get +100 tutorials for free
Follow us on:
Instagram: bit.ly/instagram-photigy
Pinterest: bit.ly/pinterest-photigy
Facebook: bit.ly/photigy-fan-page
Join on our group: bit.ly/photigy-FB-group
this is amazing I thought I was the only one using mid journey to come up with jewelery designs
That’s great video,thank you for sharing 👌
Glad you liked it! Thanks!
You are the best Alex!
Thank you, Irene!
Great for creating adds and something to learn as a hobby or business.
Absolutely!
Thank you for this excellant demonstration. I was hesitant about getting into AI but it's growing on me. Your tutuorial has inspired me to go further.
Glad you liked it. Thank you!
I love you Alex!!!
Thank you 🙏
Thank you very much!
You're welcome!
Amazing video man, Im just starting watching AI videos to undrestand, I am a portrait photographer and want to come out with some AI images for my portfilio.
Thanks!
We really appreciate your support. Thanks!
Thanks Alex, Great video!. It's still the fact that images from MidJourney are at a low resolution (max default 1024, and max 2048pix),
But, it's a great tool and great video explanation by you. It gave more insights again and that will help me forward for future product shot creativity.
It gives me directions of getting correct lighting setups and color combinations for my products shoots in a much faster way and will provide me more possible variations.
Yes, its relatively low. But with upscaler you can print them relatively large..
@@photigy Ok, Not sure if I know how it works and if I missed it, but what is large and what resolutions are involved? Or do you mean upscaling with external 3th party tools?
@@WilbertMaximus Photoshop has a neural filter that will allow you to upscale, called Super Zoom. I upscaled an AI image & printed it yesterday and it looked great. It was 12x18 inches @ 300 dpi. Anything larger than that the quality starts to degrade, but I'm really particular about printing & quality so others might upscale larger.
If you're just doing it for web than you can upscale much larger at 72 dpi
@@desireepfeiffer On the web the only thing that matters is pixel dimension e.g., the long side pixel dimension. For print, yes the dpi matters. Here is a great article on this, I can't post a url without it being removed, so just google this text, and you will find it if you're interested. "Let’s clarify this once and for all:
only pixel dimensions matter on the web"
Do u think all the backgrounds would be created in Midjourney and u would shoot product and create composite images by merging the two in Photoshop for final outcomes?
Молодец! Очень профессионально записаный ролик!
thank you
Alex, could I show the command that I entered to the IP
How do you get to see all angels of the image (product) if I wanted to get it manufactured??
Thanks. However. How do I only change one thing in the AI created image? For example. Only take away the wrinkles of the background to your image created from your stilleben.
Great video!!! One tip: you can set quality to high (--q 2 ) which means that all 4 images are created already with the big size 1024 - then if you want just one separately, the U1-U4 commands are instantaneous. But I usually just save the 4-image collage, it's 2048x2048.
Thanks for the info!
@@photigy I realized that this only happens with version 5 (--v 5). By default it is on version 4 and in that case the 4 images are not pre-made in advance.
You're all over the place, but I really got a lot out of this video.
Hi Alex, one question in my mind is, isn't AI killing professional Photographers job coz now anyone can take pictures with their phones and generate it the way they want. Don't you think brands will go AI instead of hiring a professional Photographer? Please share your thoughts.
No, I don't think so, because you cannot easily replicate images and do small changes and improvements.
The current resolution is also to small for real printed photo's.
You also cannot make series in a way that different products are in the same set-up (pack shots) for a uniform look on eg websites
I do not doubt it will improve, but I don't mind. It will create new kind of work; so called prompt writers. You need to "imagine" what you want to get out and AI is giving you suggestions.
It is not the other way around.
I use it for moodboards, idea generations and for directions of shoots.
I'm currently playing around with the tool to see how it helps me to create photo ideas and lighting ideas.
I see it as a tool that helps me, not one that destroys. 🙂
Technology will never replace a human
Agree with Wilbert - its not going to happen soon. AI can't generate an exact image of anything... IIt will DREAM about it, and the result will be like a dream about. Commercial photographers takes exactly what and how client wants:-)
Alex, could I show the command that I entered in artificial intelligence for a combination of images
Sure, what the command did you enter?
Is it free and how secure is it?
Its free. Its public - all the security is under Discord account (its sort of a messenger), and MJ has a private mode when results won't be shared on public.
@@photigy You are ridiculous how you keep removing my comment. If you have an issue with what I said, then tell me where I'm wrong. This is NOT a true statement. "Its free. Its public - all the security is under Discord account (its sort of a messenger), and MJ has a private mode when results won't be shared on public."
Seems to me that everyone loves "free money" until they realize what inflation is. Everyone loves "free images" until they realize what inflation is. In other words... There is no such thing as a free lunch.
Why you think it's a free money: Where is the money there?;)
I see it as a tool - and like any tool, it can be used to make money, or do something else, like hobby, etc.
@@photigy Its a tool, sure. My point was that much like anything, when you inflate number of something you decrease its rarity, it becomes worth...less. Smartphones have made easy for anyone to make images, and social media has made it easy to publish. This is inflation in action. Now even best images, had a hard time getting the attention because people have become so used to seeing images all the time, they really have a short attention span and less appreciation for it. AI generative imagery allows anyone with virtually no skill to create all kinds of images. It inflates the number of images in circulation, resulting in more and more images, chasing same amount of people and fewer and fewer minutes of attention span. In a few years, we will all notice the penalty of this.
Just like trillions of dollars created out of thin air during last few years. Well now all those dollars are worth less and less because they are chasing same or fewer number of goods and services. When goverment was giving out stimulus checks, people didn't complain, but now when you go buy gas for your car or you go pay your bills or buy groceries you can feel the effects of inflation, can't you? That is how it works. We can expect the effect of inflation of AI generated images to be felt once the honeymoon phase is over. That was my point. There is danger in this technology, not enough people are talking about because they are too busy plying with the new toy. In a year or two, we can revisit this conversation to see if I was right or not. In the meantime, yeah, why not, lets play. What could possibly go wrong?
Cheers!
But images are really free. Stupid @krumoslav stifter
In a year, anybody can take a photo of any product and give it any context - and AI will generate a beautiful shot with the actual product, not some fantasy mockup. No studio. No lights. No professional photographer. Portraits? Cellphone plus AI will yield better results than any studio session. Fashion? Put the clothes on a mannequin or spread them out, take some iphone shots. Next step: AI puts them onto photoreal models and the setting is Acapulco Beach or Paris. Most photographers do not get it yet. It's over, folks.
Hope that you are aware that all “ai” image generators are capable of generating images due of massive art stealing, mid journey used without consent of anyone the drawings, paintings and photographs of millions of humans in order to get its predictive algorithm trained, really sad to see someone I’ve admired and considered the best product photographer in the world promoting this stuff
Please elaborate on "stealing".. How did they do it?
@@photigy Well you should start reading about it if you don't know. Plenty to find online and it is a problem bc it takes images from anybody and pays no royalties to nobody. Ask a stock photographer..lol...I think it is pretty self explanatory!
@@photigy sure, for example mid journey, in order for it to be capable to simulate the art style of recognized human artists it’s database was feed with all paintings, drawings or photographs of that artist and it happened illegally without any consent or compensation of any kind, if you input for example the name of Nekro, Santiago Caruso, Roberto Ferri or any alive famous artist as part of your prompt the mid journey algorithm will deliver you an image that resembles any of those artists stiles and that’s because they’re developers added dozen of millions of art pieces, during the video Alex uploaded a ring photograph he took during one of his courses, now that picture became part of the mid journey database, he gave them without knowing part of his hard work for free, now the algorithm can use it to copy form it whenever someone else requires for a ring photorealistic image
@@photigy all predictive algorithms work the same way, for example chat gpt, if it’s capable of deliver specific and logical responses it’s because it’s database was feed with almost every ebook, digital journal and news website available on the web, again without consent of they’re respective authors, everyone seems to be so amazed with “ai” generators but not everyone seems to know all the illegal stuff they’re developers are doing to make it work
Unsubscribed because Alex apparently doesn't respect someone pointing out when he is wrong.
Hm.. Where it was?