Your photos reveal a lot of private information. In this experiment, we use Google Vision API to extract the story behind a single photo. See what they see.
YES! I am 77 y.o. geek with a camera in my MinisForum MC630(?)and a box of vacuum tubes for audio somewhere near this location. I have no worries about anything except what I may receive for XMAS. My shopping is complete.
The subtle wrinkles around his eyes and forehead suggest he's lived a life filled with laughter and experiences. The color and texture of his skin indicate that he spends a moderate amount of time outdoors. His expression is welcoming and open, suggesting a friendly and approachable demeanor.
Wow, super cool & fascinating! Potentially scary but if everything of that nature ruled my life these days I'd never leave the house.
I found it interesting that it came up with lots of generic info about each photo I tried, but failed to identify famous people by name. I tried a pic of Marilyn Monroe posed on a studio lot with Betty Grable & Lauren Bacall, and the closest it came was, "The setting suggests a film studio or a location where films are shot, perhaps in Hollywood, California. The lighting is bright, suggesting it might be midday. The three women appear to be actresses or entertainers."
Pretty good, but if it can't (or chooses not to) identify a face as famous as Marilyn's. . . ?
Dang! I got 3 large paragraphs describing my "Brad Ramone"/Ramones photo that I use on my FRONT ROW & BACKSTAGE front page (the tiny round thumbnail to your left, fellow readers...it also serves as my business card).
Here's the first paragraph: "The photo seems to have been taken in what seems to be a backstage area, possibly at a rock concert or music venue. The background is somewhat blurry but shows a wall with some writing on it, suggesting a less-than-formal setting. In the foreground, four individuals are sharply in focus; they appear to be a band or group of friends, given their similar styles.
"Each person is wearing casual clothing, suitable for an informal setting, and they seem to be posing for the camera. There's a sense of camaraderie among them, and the overall mood seems relaxed and informal. The background also includes an advertisement for Brad Kyle’s Substack page, indicating that the image may be related to his content."
Well, that was fun - I tried it with several of my illustrations, and they got the idea behind them pretty close. Now I have a nice tool to write my descriptions for my websites - they got a quite a few things mixed up, but that's OK. It's a good start. Thanks!
I suppose it depends on the photo you use. A Paris cityscape produced the same kind of AI drivel that turns up on almost every travel-related blog I see. 😊
YES! I am 77 y.o. geek with a camera in my MinisForum MC630(?)and a box of vacuum tubes for audio somewhere near this location. I have no worries about anything except what I may receive for XMAS. My shopping is complete.
👍
Stop your making me blush ☺️
---
The subtle wrinkles around his eyes and forehead suggest he's lived a life filled with laughter and experiences. The color and texture of his skin indicate that he spends a moderate amount of time outdoors. His expression is welcoming and open, suggesting a friendly and approachable demeanor.
Nice! Your medal, sir: 🥇
I used the same photo at the end of my newsletter. It said I was "a middle-aged man" between 50 and 60. Flattering!
Nice!
Wow, it was pretty much spot-on with the photos I used.
Thanks, Jim. It got mine bang on - from photos from my phone.
This is scary what that can do. It got a few things wrong on the photos I used.
Yes, this is a problem with this high tech stuff. Thanks for reading, Matt!
Wow, super cool & fascinating! Potentially scary but if everything of that nature ruled my life these days I'd never leave the house.
I found it interesting that it came up with lots of generic info about each photo I tried, but failed to identify famous people by name. I tried a pic of Marilyn Monroe posed on a studio lot with Betty Grable & Lauren Bacall, and the closest it came was, "The setting suggests a film studio or a location where films are shot, perhaps in Hollywood, California. The lighting is bright, suggesting it might be midday. The three women appear to be actresses or entertainers."
Pretty good, but if it can't (or chooses not to) identify a face as famous as Marilyn's. . . ?
Just like you said, scary but we can't live in a cave! Thanks for commenting!
Very cool (and a little scary?)
sure is!
Dang! I got 3 large paragraphs describing my "Brad Ramone"/Ramones photo that I use on my FRONT ROW & BACKSTAGE front page (the tiny round thumbnail to your left, fellow readers...it also serves as my business card).
Here's the first paragraph: "The photo seems to have been taken in what seems to be a backstage area, possibly at a rock concert or music venue. The background is somewhat blurry but shows a wall with some writing on it, suggesting a less-than-formal setting. In the foreground, four individuals are sharply in focus; they appear to be a band or group of friends, given their similar styles.
"Each person is wearing casual clothing, suitable for an informal setting, and they seem to be posing for the camera. There's a sense of camaraderie among them, and the overall mood seems relaxed and informal. The background also includes an advertisement for Brad Kyle’s Substack page, indicating that the image may be related to his content."
To the last sentence, I say, "Yeah, well, duh!"
That seemed to work. Cool!
Well, that was fun - I tried it with several of my illustrations, and they got the idea behind them pretty close. Now I have a nice tool to write my descriptions for my websites - they got a quite a few things mixed up, but that's OK. It's a good start. Thanks!
Glad it worked as expected for you. There is a lot of different opinions about this. Thanks!
I suppose it depends on the photo you use. A Paris cityscape produced the same kind of AI drivel that turns up on almost every travel-related blog I see. 😊
Same with text to photo software, ¯\_(ツ)_/¯
…eek…