View allAll Photos Tagged GenerativeAI
See my albums list for some of my best work: www.flickr.com/photos/200044612@N04/albums/
See my main account for my photography, videos, fractal images and more here: www.flickr.com/photos/josh-rokman/
Made with Image Creator from Microsoft Designer, formerly known as the Bing Image Creator. Powered by DALL·E 3.
I think that AI image generation is similar in many ways to photography. The camera itself handles all the fine details, but the photographer is in charge of curating the types of images that will be created.
Ultimately, it is all about maximizing the probability that something good will be created.
This is very similar to AI image generation, in terms of the skills involved and what the human does vs. what the machine does.
You can't compare AI image generation to the process of actually making these images from scratch with 3D software or paint/pencils, where the human controls every detail.
However, I think the process really is very similar to that of photography, as I made the case for above. I think that DALL-E 3 is by far the most powerful AI image generation tool currently available.
- Josh
Link to my album with the original high-definition images: www.flickr.com/photos/200044612@N04/albums/72177720324216...
The slightly longer youtube version is here: www.youtube.com/watch?v=4GYgGPX5PEI&t=10s
These are all my own images that I made with the Bing Image Creator, powered by DALL-E 3. DALL-E 3 (as implemented by the Bing Image Creator) has gotten worse since these were made - it may be difficult or impossible to create images of this quality today.
The song in the video is my own original creation - it is NOT an AI generated song.
Many of the images in this video slideshow could not be achieved through real photography. Extreme micro photography (normally referred to as ‘macro photography’) creates an image with a very shallow depth of field, meaning that only a small sliver of the image will be in focus.
To compensate, a technique called 'focus stacking' is used, where numerous photos are taken with slightly different focus points, and computer software then merges the sharpest parts of the images together into one image. The process of taking these different photos for focus stacking takes some time, and so it will not work for moving objects, such as moving insects. A camera may need a shutter speed as fast as 1/500 of a second to sharply capture images such as this – but focus stacking may require 50+ shots to be taken at different focal points - without the subject moving.
The insects photographed with this technique are usually no longer alive, or at least they are keeping still for a long period of time. Only with AI is it possible to capture extreme close-up images of moving insects where more than a sliver of the image is in focus - with the realism level of a real photo. My goal with AI is to create images that would have been impossible to create without it.
- Josh
I grabbed this fairly boring photo of the moon this morning while out with the dog and figured it would be a good shot to practice with the AI tool in Photoshop. It was a lot fun so I'll have to try and remember to do this a little more often.
Hope everyone is doing well.
Click "L" to view on black.
See my albums list for some of my best work: www.flickr.com/photos/200044612@N04/albums/
See my main account for my photography, videos, fractal images and more here: www.flickr.com/photos/josh-rokman/
Made with Image Creator from Microsoft Designer, formerly known as the Bing Image Creator. Powered by DALL·E 3.
I think that AI image generation is similar in many ways to photography. The camera itself handles all the fine details, but the photographer is in charge of curating the types of images that will be created.
Ultimately, it is all about maximizing the probability that something good will be created.
This is very similar to AI image generation, in terms of the skills involved and what the human does vs. what the machine does.
You can't compare AI image generation to the process of actually making these images from scratch with 3D software or paint/pencils, where the human controls every detail.
However, I think the process really is very similar to that of photography, as I made the case for above. I think that DALL-E 3 is by far the most powerful AI image generation tool currently available.
- Josh
Photo of a vintage vase expanded and manipulated in an image created with AI using Firefly, and Photoshop.
Created in Leonardo AI.
"May The Force Be With You"
See more here: www.youtube.com/@journeymanplayer7459
With Adobe Photoshop adding generative AI to its software, one will never quite know what in a photo is real or not anymore. Granted, you'd be suspicious of many of these - but this is the beta version. In six months it'll be largely flawless. The image has embedded metadata that indicates it was AI modified, but not many people know how to look at the metadata. To make these objects I simply chose a space and typed in the object I wanted. How many AI created objects can you find? Or maybe easier, what's real in here?
Well, not really a sunset. The sun has been added to the image with the help of Photoshop's artificial intelligence. The ship and the waves are real. The photo was taken on the beach at Yport, a popular holiday destination on France's Channel coast. It was around midday, but dense high fog created a rather dim, diffuse lighting.
Here is a link to the photo without additions:
www.flickr.com/gp/bernd_thaller/6s0449Z75x
I'm not sure I want to make changes like this more often. In any case, it no longer corresponds to my own memory of this scene. But I'm still impressed with how Photoshop rendered the sun along with atmospheric effects and waves on the horizon.
Created in DALL-E 3.
This one reminds me of some of the better artwork featured in the old, "Choose Your Own Adventure" books, or even the spaceship scene in the classic PC adventure game, "Space Quest I: The Sarien Encounter," where your tiny ship is approaching the Sariens feared, "Deltaur" capitol ship.
See more here: www.youtube.com/@journeymanplayer7459
See my main account for my photography, videos, fractal images and more here: www.flickr.com/photos/josh-rokman/
Made with Image Creator from Microsoft Designer, formerly known as the Bing Image Creator. Powered by DALL·E 3.
I think that AI image generation is similar in many ways to photography. The camera itself handles all the fine details, but the photographer is in charge of curating the types of images that will be created.
Ultimately, it is all about maximizing the probability that something good will be created.
This is very similar to AI image generation, in terms of the skills involved and what the human does vs. what the machine does.
You can't compare AI image generation to the process of actually making these images from scratch with 3D software or paint/pencils, where the human controls every detail.
However, I think the process really is very similar to that of photography, as I made the case for above. I think that DALL-E 3 is by far the most powerful AI image generation tool currently available.
- Josh
In real life, that barn is red, but there are elements in this photo that were "helped" by AI. Can you guess which three?
I am very impressed by the level of photo-realism in this one.
See my main account for my photography, videos, fractal images and more here: www.flickr.com/photos/josh-rokman/
Made with Image Creator from Microsoft Designer, formerly known as the Bing Image Creator. Powered by DALL·E 3.
I think that AI image generation is similar in many ways to photography. The camera itself handles all the fine details, but the photographer is in charge of curating the types of images that will be created.
Ultimately, it is all about maximizing the probability that something good will be created.
This is very similar to AI image generation, in terms of the skills involved and what the human does vs. what the machine does.
You can't compare AI image generation to the process of actually making these images from scratch with 3D software or paint/pencils, where the human controls every detail.
However, I think the process really is very similar to that of photography, as I made the case for above. I think that DALL-E 3 is by far the most powerful AI image generation tool currently available.
- Josh