Sean Connery standing next to a TVR Tuscan...
Discussion
I know it's a random title to post...basically, one of the forum members here has doctored the famous picture of Sean Connery standing next to his DB5 on the Furka Pass in Goldfinger and replaced the DB5 with a TVR Tuscan 
I am trying to locate the member so if you are him or know him could you message me. I was kindly given a digital copy of the image from which I had a mug designed...sadly I broke the mug so hoped to replace it but cannot find the pic....It's a long shot I know but thanks for reading this...
I am trying to locate the member so if you are him or know him could you message me. I was kindly given a digital copy of the image from which I had a mug designed...sadly I broke the mug so hoped to replace it but cannot find the pic....It's a long shot I know but thanks for reading this...Fimbo said:
Wow, ok, so thats pretty amazing!! I may be about to embarrass myself asking, but what/how did you ask chatgpt to produce this? Thanks so much for your help.
No worries. I took a screen shot of a Google image of Sean Connery and a side on picture of my Tuscan and uploaded both to ChatGPT. I then asked it to replace the original car in the mountain picture with the tvr in the other image.I then had a couple of tweets, as it put the car around the wrong way and stuff but you can simply ask it more questions. For example I asked to change the car to a lighter colour, it asked me a question back about what colour I actually wanted (including if I wanted Aston Martin’s silver…which I thought was clever) and it came back with this.
I also gave it a picture of my grantura and it did this
porterpainter said:
GAjon said:
Me thinks you expect too much for 5 mins of AI time on a free program.Don t think it ll be noticeable when shrunk to mug size
Not that I am being pedantic.
This prompted my first ever try at using ChatGPT. There are clearly limits to what it's allowed to do :
" Why I couldn't generate the image :
You asked for : Add Batman leaning on this car. Batman is a copyrighted character, and adding a specific copyrighted character into a real photo is considered generating non-transformative copyrighted content, which I m not allowed to produce. This rule covers:
Putting Batman directly into real photos
Creating realistic Batman images
Editing photos to add Batman in a way meant to replicate the copyrighted character
- Why did I allow the 3D cartoon earlier?
Because that transformation produced a new stylized, fictional, non-realistic cartoon version of the people - a transformative reinterpretation. Adding Batman into your real car photo wouldn t be transformative enough - it would be reproducing a specific trademarked/copyrighted character as-is.
- What I can do instead
I can add:
A generic superhero inspired by Batman s style
A dark-cape vigilante character
A silhouette or symbolic figure
A parody version clearly not Batman (different costume, emblem, colors) "
This is the best I could get it to do, with the last of those options :
Interestingly, as with the images of Connery, the AI has apparently re-drawn the whole car as a close approximation of my original photo - the wheels, door-mirrors, front indicators and spoiler aren't correct. But after two tries, it then told me there were limits on how many images it allows one person to make, and to be fair to other users, I had to wait an hour before trying again ...
" Why I couldn't generate the image :
You asked for : Add Batman leaning on this car. Batman is a copyrighted character, and adding a specific copyrighted character into a real photo is considered generating non-transformative copyrighted content, which I m not allowed to produce. This rule covers:
Putting Batman directly into real photos
Creating realistic Batman images
Editing photos to add Batman in a way meant to replicate the copyrighted character
- Why did I allow the 3D cartoon earlier?
Because that transformation produced a new stylized, fictional, non-realistic cartoon version of the people - a transformative reinterpretation. Adding Batman into your real car photo wouldn t be transformative enough - it would be reproducing a specific trademarked/copyrighted character as-is.
- What I can do instead
I can add:
A generic superhero inspired by Batman s style
A dark-cape vigilante character
A silhouette or symbolic figure
A parody version clearly not Batman (different costume, emblem, colors) "
This is the best I could get it to do, with the last of those options :
Interestingly, as with the images of Connery, the AI has apparently re-drawn the whole car as a close approximation of my original photo - the wheels, door-mirrors, front indicators and spoiler aren't correct. But after two tries, it then told me there were limits on how many images it allows one person to make, and to be fair to other users, I had to wait an hour before trying again ...
Edited by frontfloater on Sunday 30th November 17:41
v8sag said:
1. It's not a Sag it's a Tam.2. It looks fantastic in white - well done that's a brave choice assuming you went for the colour change.
3. Your blow-up Barbie is clearly the XXL deluxe model at 7' 2". I just hope she has sufficient airbags to cope with a full frontal collision or else she's going to lose her head on the rollcage inside the top of the windscreen.
Great thread guys.
Any more?
harryt said:
v8sag said:
1. It's not a Sag it's a Tam.2. It looks fantastic in white - well done that's a brave choice assuming you went for the colour change.
3. Your blow-up Barbie is clearly the XXL deluxe model at 7' 2". I just hope she has sufficient airbags to cope with a full frontal collision or else she's going to lose her head on the rollcage inside the top of the windscreen.
Great thread guys.
Any more?
Gassing Station | General TVR Stuff & Gossip | Top of Page | What's New | My Stuff