All these recent posts from Krantzbucks have inspired me to go back and take another look at photo generative ai apps. I have to admit they are getting scary good...and I have mixed emotions about that.
Photos are from a session done with Google Gemini (a.k.a. nano banana).
Original below was cropped from an offering at historicalimages.com.
Prompt:
Restore and colorize the photo
Which gave me this:
Past experience has shown that these apps can alter facial features. On this one it didn't do too badly. A nice bonus is that the watermark disappeared without any prompting.
Since I wanted Rautzhan with a cap, I uploaded this from ebay:
Prompt:
add this cap to the ballplayer
And voila...I got this:
I do note the aspect ratio changed (and generated appropriate pixels left and right). I think that is because the cap was uploaded after the original player image so the app used the aspect ratio of the cap image.
Finally, knowing that Rautzhan sported a mustache at some point that year, I did one more prompt to allow for that option.
Prompt:
add a mustache to the player
Which gave this final image:
I don't believe for a minute that every effort will work as well as this one did but the technology is only going to get better and, I believe, quite rapidly...and as previously mentioned that can be good or bad depending on how it is used.