Something else you can try is an AI video creation software like Hailuo. I have played with it a little. It has free as well as paid accounts. I paid $10 a month for 2 months and switched back to free. It has some weird rules on account usage and credits that annoyed me enough to stop paying. Anyway, you can create 6 second videos with paid or free accounts. There are two differences in the free and paid accounts, a paid account has no watermarks on the final video and they generate in a few minutes as opposed to 20 to 40 minutes per video on the free account.
To create a video you can tell the AI what you want and let it create a video from scratch. Another option is to upload an image and tell it what you want to happen. I have been using both, but uploading an image works better in my opinion.
FYI, Hailuo is located in China, but I had no problems using it or canceling my paid subscription. It understands English very well.
If you do create an account you get 1,000 free credits (it takes 30 credits to create each video). You only have 3 days to use these free credits. You additionally get 100 free credits a day with either free or paid accounts, but they expire every 24 hours.
If people are interested in using Hailuo I can give you some helpful hints on how it works.
Has anyone used ChatGPT to create images of themselves as Batman, Robin, etc.?
Moderators: Scott Sebring, Ben Bentley
Re: Has anyone used ChatGPT to create images of themselves as Batman, Robin, etc.?
Also very informative, Dell!
-
njboywonder
- Posts: 0
- Joined: Mon Sep 03, 2012 6:10 am
Re: Has anyone used ChatGPT to create images of themselves as Batman, Robin, etc.?
Anything you can offer will be appreciated. Would love to create video.
Getting photo images suitable for posting here is proving frustrating. The more TV series-centric, the tougher time it has due to all the different request details and wording, elements, accumulated knowledge, and then just aesthetics. I'm doing well with images recreating more straightforward experiences I've had in costume/photos I've had taken--using a single photo image of me as Robin and a simple request so the program doesn't have to reference a ton of Batman information.
I've been putting notes into a post as I've been going along, but was waiting to post another image with them. Instead, I'll post what I've learned now.
BTW, how good are the videos??
Thanks, Dell!
Getting photo images suitable for posting here is proving frustrating. The more TV series-centric, the tougher time it has due to all the different request details and wording, elements, accumulated knowledge, and then just aesthetics. I'm doing well with images recreating more straightforward experiences I've had in costume/photos I've had taken--using a single photo image of me as Robin and a simple request so the program doesn't have to reference a ton of Batman information.
I've been putting notes into a post as I've been going along, but was waiting to post another image with them. Instead, I'll post what I've learned now.
BTW, how good are the videos??
Thanks, Dell!
-
njboywonder
- Posts: 0
- Joined: Mon Sep 03, 2012 6:10 am
Re: Has anyone used ChatGPT to create images of themselves as Batman, Robin, etc.?
More that I've learned...
To give you an idea, I've made many dozens of attempts since my last post and not gotten anything that was exactly what I wanted. I probably tried about 20 times to have it redo the image of me on the Batpole. Since I'm colorblind, I can't tell if it captured Robin's tights or not in that one (Anyone?)--or if there's even a definitive answer to that.
If you give it specifics, it sometimes struggles. Usually it gives what you want but just not in an ideal way. But sometimes I just, for instance, don't like what it does with my face.
I can, though, ask it to put me as Robin in a trap with Batman and it will generate its own ideas. Many of these are great to see, fun, but not what I'm looking for in terms of something exacting that I can print and hang.
It won't accept certain words. When it depicts Robin without tights, and I indicate to add them, a few times it made them green. When I instructed to make them flesh colored, it refused, (indicating it suggested nudity). It rejects words or phrases like tied up or suspended, but will not only create those images with different words, but will sometimes offer them itself if you enter a request that violates its policies (in virtually all cases, unlike with the tights note above, not indicating what the policy/violation is (though you can assume and otherwise learn quickly enough)). I get it but it's frustrating to waste so much time, effort, and allotted (paid) attempts.
The good thing is that it is susceptible to suggestion, so Robin tied to a pole might not work, but Robin against a pole with his hands behind his back not only will, but will result in the image including rope.
Also frustrating, sometimes it's clearly an AI image. A very few times it looks like a real photo, which is what I want. No rhyme or reason as to how to force it as the outcomes don't always correspond with the request. But when it does what you like once, best thing is to continue working with the initial successful result as it will replicate its image style. Similarly, if it gives a few results that are undesirable in the same way, move to another project and restart.
As I start asking it to recreate some experiences I've had, I find that requests involving a contemporary outdoor scene, like a park, produce more photorealistic results. I assume it's association with the TV series' classic indoor settings like warehouses suggest slightly more stylization, even if instructed otherwise.
To give you an idea, I've made many dozens of attempts since my last post and not gotten anything that was exactly what I wanted. I probably tried about 20 times to have it redo the image of me on the Batpole. Since I'm colorblind, I can't tell if it captured Robin's tights or not in that one (Anyone?)--or if there's even a definitive answer to that.
If you give it specifics, it sometimes struggles. Usually it gives what you want but just not in an ideal way. But sometimes I just, for instance, don't like what it does with my face.
I can, though, ask it to put me as Robin in a trap with Batman and it will generate its own ideas. Many of these are great to see, fun, but not what I'm looking for in terms of something exacting that I can print and hang.
It won't accept certain words. When it depicts Robin without tights, and I indicate to add them, a few times it made them green. When I instructed to make them flesh colored, it refused, (indicating it suggested nudity). It rejects words or phrases like tied up or suspended, but will not only create those images with different words, but will sometimes offer them itself if you enter a request that violates its policies (in virtually all cases, unlike with the tights note above, not indicating what the policy/violation is (though you can assume and otherwise learn quickly enough)). I get it but it's frustrating to waste so much time, effort, and allotted (paid) attempts.
The good thing is that it is susceptible to suggestion, so Robin tied to a pole might not work, but Robin against a pole with his hands behind his back not only will, but will result in the image including rope.
Also frustrating, sometimes it's clearly an AI image. A very few times it looks like a real photo, which is what I want. No rhyme or reason as to how to force it as the outcomes don't always correspond with the request. But when it does what you like once, best thing is to continue working with the initial successful result as it will replicate its image style. Similarly, if it gives a few results that are undesirable in the same way, move to another project and restart.
As I start asking it to recreate some experiences I've had, I find that requests involving a contemporary outdoor scene, like a park, produce more photorealistic results. I assume it's association with the TV series' classic indoor settings like warehouses suggest slightly more stylization, even if instructed otherwise.
Re: Has anyone used ChatGPT to create images of themselves as Batman, Robin, etc.?
Several tips for creating short videos from images:
1) Use uncluttered images if possible:
The AI will "look" at the image and try to interpret what is there. Here is a real example. I used a still of Leslie Gore as Pussycat from "That Darn Catwoman". I erased Aunt Harriet from behind Leslie as I assumed the AI wouldn't know what to do with the small amount of Aunt Harriet that could be seen. So I basically turned Aunt Harriet into a yellow pillow. Alfred's green smock can be seen, but it didn't look like a person so I just left it alone and that worked ok. I asked the AI to have her turn and walk (6 seconds is pretty short so don't ask for much).
Pussycat had a pair of mittens that she wore on a string and had them hanging around her neck behind her. Now we know what it looked like, but the AI guessed from the photo. It was a reasonable guess, but quite wrong.
Sorry, images don't seem to be working.
(no Aunt Harriet)
(mittens?)
2) Be careful with your wording:
I used a still of Julie from a Catwoman episode and said have her raise her hands. In the still she was wearing her gloves, but as she raised her hands the gloves just vanished, and her hands were bare.
1) Use uncluttered images if possible:
The AI will "look" at the image and try to interpret what is there. Here is a real example. I used a still of Leslie Gore as Pussycat from "That Darn Catwoman". I erased Aunt Harriet from behind Leslie as I assumed the AI wouldn't know what to do with the small amount of Aunt Harriet that could be seen. So I basically turned Aunt Harriet into a yellow pillow. Alfred's green smock can be seen, but it didn't look like a person so I just left it alone and that worked ok. I asked the AI to have her turn and walk (6 seconds is pretty short so don't ask for much).
Pussycat had a pair of mittens that she wore on a string and had them hanging around her neck behind her. Now we know what it looked like, but the AI guessed from the photo. It was a reasonable guess, but quite wrong.
Sorry, images don't seem to be working.
(no Aunt Harriet) (mittens?)2) Be careful with your wording:
I used a still of Julie from a Catwoman episode and said have her raise her hands. In the still she was wearing her gloves, but as she raised her hands the gloves just vanished, and her hands were bare.
dell
-
njboywonder
- Posts: 0
- Joined: Mon Sep 03, 2012 6:10 am
Re: Has anyone used ChatGPT to create images of themselves as Batman, Robin, etc.?
One extremely cool thing is that you can use the photos created from Chat GPT to create videos on HAiluo. Trying that now.
Re: Has anyone used ChatGPT to create images of themselves as Batman, Robin, etc.?
I imagine most of this will improve within a short amount of time. Probably by the end of the year it will be more intuitive and accurate in producing what you want.
Brett, once you get an image of you that you like, consider using that software that Randy has been posting about to create/print your very own Marx Toys style figure!
Brett, once you get an image of you that you like, consider using that software that Randy has been posting about to create/print your very own Marx Toys style figure!
-
njboywonder
- Posts: 0
- Joined: Mon Sep 03, 2012 6:10 am
Re: Has anyone used ChatGPT to create images of themselves as Batman, Robin, etc.?
Wow. No idea on that one, BP. Will search here soon enough. Thanks.
-
njboywonder
- Posts: 0
- Joined: Mon Sep 03, 2012 6:10 am
-
njboywonder
- Posts: 0
- Joined: Mon Sep 03, 2012 6:10 am
Re: Has anyone used ChatGPT to create images of themselves as Batman, Robin, etc.?
It's amazing what you can do by combining the two sites. I can recreate experiences I had in real life (scenes I recreated) extremely realistically.
Re: Has anyone used ChatGPT to create images of themselves as Batman, Robin, etc.?
njboywonder wrote: ↑Sun Jun 15, 2025 12:29 pm It's amazing what you can do by combining the two sites. I can recreate experiences I had in real life (scenes I recreated) extremely realistically.
Somehow you have to post those videos so that we can see what that looks like!
-
njboywonder
- Posts: 0
- Joined: Mon Sep 03, 2012 6:10 am
Re: Has anyone used ChatGPT to create images of themselves as Batman, Robin, etc.?
I can't post here. Thinking about YouTube.
One big thing I'm learning regarding the video is that it works pretty well with photorealistic images but adding props that would be involved in the action--such as a net--is not handled well.
Adding a setting--such as directing the action toward an off screen warehouse door--works pretty well.
One big thing I'm learning regarding the video is that it works pretty well with photorealistic images but adding props that would be involved in the action--such as a net--is not handled well.
Adding a setting--such as directing the action toward an off screen warehouse door--works pretty well.
-
njboywonder
- Posts: 0
- Joined: Mon Sep 03, 2012 6:10 am
Re: Has anyone used ChatGPT to create images of themselves as Batman, Robin, etc.?
That is incredible! Just imagine where this will all be in five or 10 years. Future generations may never see real actors on film again. I can imagine once popular actors and their estates will be licensing likenesses. Perhaps there will be an all new genre of popular AI actors.