Truly a miraculous convenience-no need for images, face swapping, or training LoRA models. If combining two characters or two face models is possible, that would be amazing!"
Did you add negative prompts while creating some of your videos cause I saw there were some inaccuracies in face lighting !! Is it due to not using proper negative prompts?
the agreement in the bottom said that you were not aloud to remove the Kling Ai watermarks. I don't see any watermarks on your videos. Did I mis-read it?
The agreement says, 'you shall not remove or tamper with the brand logo or you shall prominently indicate that the Output is generated by "KLING AI" in the usage scenarios of the generated content.' This is a Kling Ai Tutorial video lol.
Awesome! I’m trying to train it with my AI character by creating pictures and then short videos to train Kling. However, I keep getting the message, "This is not the same person"-so frustrating! :)
Thanks for sharing this, I was interested in giving this a try, but I wasn't confident on how the results would turn out. How are the outputs from the Model that you trained? Also, what kind of Image Style was the character. :)
Mine worked but I used the same Midjournery image to generate 10 different Kling videos first, tell her to look up and left and right, and wave hand and such.
This is an amazing vid, unfortunately my model completed 6 and a half hours and still not finished , and what's making me worried is that when i enter to check it's saying something like hold on and the generate button is still on , it's like I can press it in the time that I already pressed it before to start ... so am afraid to press generate and then it goes from zero again !! can you advise pls ? ( I contacted kling team support but unfortunately no answer till now !!! Note : They deduct the 999 credits already
Don’t worry I had a similar issue and wasn’t sure if it was working. If it’s been deducted from your credits, I would say it’s working. It may be a lot of traffic. Hold off and wait for email. Worst case scenario email them.
what about the error: Failed due to unknown error, im trying to upload each video, im doing everything perfect, format, lighting, poses, any of you have had this error? im using chrome, could that be it?
Hey dude, I had a look of errors whilst doing the same thing aswell. One thing that worked for me was uploading one video at a time. I was using edge, so check to see if that works. Also I noticed the videos that kept failing is when I covered my face or turned my head away.
@ the issue is i was moving alot, too far awsy from camera, its better if it is from the shoulder up, and i was using Capcut to export the video!!! If you use Final Cut pro or adobe premier pro it will work!!!
Truly a miraculous convenience-no need for images, face swapping, or training LoRA models. If combining two characters or two face models is possible, that would be amazing!"
OMG, man, this tutorial is the best and most awesome as far as consistency features are concerned, explained so meticulously and in detail.
Thank you and I appreciate you watching.
Did you add negative prompts while creating some of your videos cause I saw there were some inaccuracies in face lighting !! Is it due to not using proper negative prompts?
Good explanation of the new feature. Thanks.
Those are nice.
Thank you for your vid not just showing half naked ai girls. Some of these creators it's all they show to get views.
Thank you, man appreciate, you watching.
You did not show how to setup the workflow
Good stuff
the agreement in the bottom said that you were not aloud to remove the Kling Ai watermarks. I don't see any watermarks on your videos. Did I mis-read it?
The agreement says, 'you shall not remove or tamper with the brand logo or you shall prominently indicate that the Output is generated by "KLING AI" in the usage scenarios of the generated content.'
This is a Kling Ai Tutorial video lol.
@@AiTechRealm Thank you sir!
Can you share some prompts of the fancy scenes? I tried this with underwater scenes and the quality wasn’t so good😅
See if you can use your face model to generate an image of you underwater, and then send it to image-to-video..? Just an idea
Ops, just checked. I don't think you can do that.
@@Desiak-d3x yeah it only supports text2video using face model...
Brill
Can I use my dog to make a custom face model in this new feature?
I don't think so but no 100%
Awesome! I’m trying to train it with my AI character by creating pictures and then short videos to train Kling. However, I keep getting the message, "This is not the same person"-so frustrating! :)
focus on lighting
@@ijazulhaqkhan187 I finally managed that kling accepted 25 videos, the results are ok but I need to work on this now :)
Thanks for sharing this, I was interested in giving this a try, but I wasn't confident on how the results would turn out.
How are the outputs from the Model that you trained? Also, what kind of Image Style was the character. :)
Mine worked but I used the same Midjournery image to generate 10 different Kling videos first, tell her to look up and left and right, and wave hand and such.
I faced the unknown error while uploading videos… have you faced the same recently?
why u need both pro and premium isnt premium or just pro not enough
This is an amazing vid,
unfortunately my model completed 6 and a half hours and still not finished , and what's making me worried is that when i enter to check it's saying something like hold on and the generate button is still on , it's like I can press it in the time that I already pressed it before to start ... so am afraid to press generate and then it goes from zero again !! can you advise pls ? ( I contacted kling team support but unfortunately no answer till now !!!
Note : They deduct the 999 credits already
Don’t worry I had a similar issue and wasn’t sure if it was working. If it’s been deducted from your credits, I would say it’s working. It may be a lot of traffic. Hold off and wait for email. Worst case scenario email them.
@@AiTechRealm Thx Buddy , I'll do that Insha'Allah ... tHX FOR WRITING BACK , APPRECIATE IT
what about the error: Failed due to unknown error, im trying to upload each video, im doing everything perfect, format, lighting, poses, any of you have had this error? im using chrome, could that be it?
Hey dude, I had a look of errors whilst doing the same thing aswell. One thing that worked for me was uploading one video at a time. I was using edge, so check to see if that works. Also I noticed the videos that kept failing is when I covered my face or turned my head away.
@@AiTechRealm thank you for the quick response!! new sub here :), ok i will try edge! i was doing one video at a time, but still getting that error,
When you get it to work come back and let us know just in case anyone else has this issue. Tell us what you did to fix it. Thanks man
i have same problem. Is anyone can solved this problem?
@ the issue is i was moving alot, too far awsy from camera, its better if it is from the shoulder up, and i was using Capcut to export the video!!! If you use Final Cut pro or adobe premier pro it will work!!!
Can I put my head on another video of a person speaking?
Pretty good. Wonder if Flux Ultra Raw trained LORAs (maybe some day?) would work even better for realism.
I feel like this is coming soon
✌️🔥
They need to increase 10 second to at least 3 minutes
most the lighting on the examples looks fake sadly. thanks for sharing though man. cheers!
thanks for watching man and i agree with you. the lighting is weird in the generations.