ADVERTISEMENT

With AI on the rise, we see more and more artwork made with the help of description-based software. It is amazing what our words can create, and it is interesting to think that good writing someday might overcome photography. Obviously, you wouldn't be able to capture true moments that happened, but you can very well create images of events and people that have never existed.

Dimitar Karanikolov, an architect, who has gained a lot of attention through his photographs, has also released a series of photographic portraits of shamans and healers from different parts of the world called Imaginary Shamans. In this series, you will see realistic-looking images of people that are not real. As Dimitar described, this is "AI travel photography".

To understand more about what goes into a description to make such detailed imagery, Bored Panda got in touch with the author. See the interview below and let us know in the comments how these pictures make you feel, Pandas.

More info: Instagram | behance.net | twitter.com | foundation.app | Facebook

Dimitar, being an architect, doesn’t let that title restrict him from other life passions. He shares what has inspired him to try and create with AI: “I’m an architect, but travel photography has been my passion for the past 8 years. I did numerous trips in order to explore and photograph authentic communities around the world: Mongolia, Ethiopia, Bolivia, Indonesia, etc.”

“I have always been fascinated by indigenous people and their culture, their rituals, and aesthetics. These are our ancestors, our roots. When AI software became wildly popular/open access several months ago, I naturally tried generating spaces and architectural details, but creating human faces and characters brings much more emotion and connection and does much more powerful visions. So I’ve decided to do some AI travel photography.”

#2

karanikolov Report

Add photo comments
POST
#3

karanikolov Report

Add photo comments
POST
ADVERTISEMENT

To get such detailed, lively, and realistic-looking images, there had to be a very descriptive text for AI to analyze. Dimitri shared his creative process with software: “The new version of the Midjourney platform is easier to work with and much more realistic as a style (if that’s what you are after). All these portraits are generated based on descriptions with words only. In order to have a more controlled result, I was very specific and described a lot of the details I wished to see in the final image - the age, the clothes, the ethnicity, etc. I’ve also specified the camera settings, the 'virtual lens' I wish to use, the light scenario, and the framing. The more words - the better!”

ADVERTISEMENT

AI is already here, and it is not going away. Therefore, we were interested to know what Dimitri thinks of this technology: “It is fascinating and scary at the same time. I understand why a lot of people feared this technology. I also work in the field of CG visualizations (www.meshroom.com), and we already lost a big project as the client decided to use AI instead of our services."

ADVERTISEMENT

Dimitar continued on AI: "Still, I think it is an amazing tool that gives lots of people the opportunity to express themselves and visualize their ideas. Surely it will have a major impact on the photographic industry in the future. Soon, when we post a photo, we’ll have to specify whether it’s real or AI, as there will be no difference in the quality!”

#8

karanikolov Report

Add photo comments
POST
ADVERTISEMENT
#9

karanikolov Report

Add photo comments
POST
#10

karanikolov Report

Add photo comments
POST
ADVERTISEMENT
#11

karanikolov Report

Add photo comments
POST
#12

karanikolov Report

Add photo comments
POST
ADVERTISEMENT
ADVERTISEMENT
See Also on Bored Panda
#19

karanikolov Report

Add photo comments
POST
#21

karanikolov Report

Add photo comments
POST
#23

karanikolov Report

Add photo comments
POST
#24

karanikolov Report

Add photo comments
POST
#25

karanikolov Report

Add photo comments
POST
#26

karanikolov Report

Add photo comments
POST
ADVERTISEMENT
See Also on Bored Panda
#27

karanikolov Report

Add photo comments
POST