Kathutet
New Member
- Apr 21, 2025
- 8
- 24
Kathutet
New Member
- Apr 21, 2025
- 8
- 24
Since I really enjoy seeing short vids of Blackadder Women getting pounded by Monsters this is my first attempt.
Kathutet
New Member
- Apr 21, 2025
- 8
- 24
Hi,
I guess this is the right thread so I'm posting some AI image-to-video work I recently created using the following prompt: "Generate a high-resolution, realistic video animation based on the uploaded image, maintaining full facial identity, structure, and expressions. The face must remain clear, centered, and consistent throughout. Use natural lighting and smooth, minimal movements to preserve the person’s recognizable features". I think the real challenge is maintaining the same high quality as Blackadder while keeping facial expressions as accurate as possible. Here is something pretty advanced:
If you have a good GPU available and are into this kind of thing, I highly recommend installing ComfyUI. It lets you do everything locally, so you won’t need to pay third-party websites for something you can handle yourself. Installing the WAN 2.1 workflow is pretty straightforward. To really bring BA sets to life, I can even imagine creating a full movie - with sound effects, cinematic cuts, and all sorts of creative touches. Maybe someday I’ll get a better computer with a decent NVIDIA graphics card to make that happen.
Happy to see if more of his work will be animated and shared!
I guess this is the right thread so I'm posting some AI image-to-video work I recently created using the following prompt: "Generate a high-resolution, realistic video animation based on the uploaded image, maintaining full facial identity, structure, and expressions. The face must remain clear, centered, and consistent throughout. Use natural lighting and smooth, minimal movements to preserve the person’s recognizable features". I think the real challenge is maintaining the same high quality as Blackadder while keeping facial expressions as accurate as possible. Here is something pretty advanced:
You must be registered to see the links
If you have a good GPU available and are into this kind of thing, I highly recommend installing ComfyUI. It lets you do everything locally, so you won’t need to pay third-party websites for something you can handle yourself. Installing the WAN 2.1 workflow is pretty straightforward. To really bring BA sets to life, I can even imagine creating a full movie - with sound effects, cinematic cuts, and all sorts of creative touches. Maybe someday I’ll get a better computer with a decent NVIDIA graphics card to make that happen.
Happy to see if more of his work will be animated and shared!
Sadly I have AMD GPU so pollo and facefusion will have todo for now.
For anyone with Nvidia 30XX and above this might be usefull
For anyone with Nvidia 30XX and above this might be usefull
You must be registered to see the links
Kathutet
New Member
- Apr 21, 2025
- 8
- 24
This was done with just 2 credits on pollo lol not worth it imo
Kathutet
New Member
- Apr 21, 2025
- 8
- 24
Kathutet
New Member
- Apr 21, 2025
- 8
- 24
You must be registered to see the links
of my take on Gisela. Like before, I'd recommend grabbing Aubrey 8 for some of the textures. Her body still needs some work, but I'm not too worried about getting an exact match. Also, sorry about the lazy render.
You must be registered to see the links
and
You must be registered to see the links
You can do it for free no GPU required.
Tried facefusion on couple of my favorite images.
bizak93
Member
- Jul 31, 2021
- 264
- 141
Piotrek
Member
- Aug 19, 2018
- 205
- 318
You could try open source WAN VIDEO i2v model. Civitai or other sites offer generation fair prices. Or can even run locally free. Many lora for animation NSFW are available to dl Civitai or huggingface tensorart etc.
Quiet132
Newbie
- Jul 21, 2017
- 79
- 508
Here are my examples of working with a local WAN I2V using the 480P model, as it has more LoRas and my VRAM is insufficient for more. I can't loop for now, so I'll try it in the future. For now, it's enough.
Here is an attempt to use the T2V model, which makes videos more realistic and loses the CG style.
Full nelson Lora is good enough for now.
Side doggy is a strange thing, sometimes it moves a woman or a man, it works at random and ignores prompts. For example, two of the same. One uses Side Doggy Lora, the other one just prompts.
Deepthroats and blowjobs work well. I have made one example of this, but I have tried many more with other 3D creators.
Here are examples of other creators.
Give it a year or two, and there will be more models with better quality.
You don't have permission to view the spoiler content.
Log in or register now.
You don't have permission to view the spoiler content.
Log in or register now.
You don't have permission to view the spoiler content.
Log in or register now.
You don't have permission to view the spoiler content.
Log in or register now.
You don't have permission to view the spoiler content.
Log in or register now.
You don't have permission to view the spoiler content.
Log in or register now.