Spit on camera LoRa for - I2V Wan 2.2 Video LoRa - K3NK
Trained on a dataset of 32 clips with various resolution, 1sec to 3sec
Various different source videos.
Agumented videos x4 to get 160 videos in total (horizontal flip, slight rotation, brightness and contrast)
The workflows im using for wan2.2 is:
* you can also find the workflow inside the sample video's metadata
Example Prompts
a woman spitting to the viewer. the composition draws attention to the mouth and jaw movements. the spitting motion appears controlled and deliberate. she spits with force and precision directly at the viewer. The spit hits the camera lens and blocks the view.a woman spitting to the viewer.***** This LoRA is shared without warranties and with the condition that it is used in a lawful and responsible way. I do not support or take responsibility for illegal, harmful, or harassing uses. By downloading or using it, you accept that you are solely responsible for how it is used. *****
Description
FAQ
Comments (15)
got it to work.. mistake on my end
It's working fine for me, I'm using this workflow: https://civitai.com/models/2188676/wan-22-i2v-extended-video
@AIXXXFunTimes it works like its not distorted, just wrong. tried different weights, different schedulers, and 3 different checkpoints, idk.... thx for the heads-up tho!
@AIXXXFunTimes are u using lightning loras or a checkpoint with lightning baked in ? i dont use comfy so workflows are not compatible.
@jynkz41 I'm using the light loras
@AIXXXFunTimes omfg it would help if i would have weighted the loras properly, man im slow, lol. i had the low lora on high phase and high lora on low phase. ben trouble shooting this for hours now! i usually run tons of loras at once and pretty complex lora strengths and combos. damn im smooth... thx again for your input =)
@jynkz41 well at least you've found out now, I've done that before - so annoying.
good job. looking forward to spitting at a subject
can we get a cum on lens lora someday?
Do you have a workflow for making Wan 2.2 loras? I'd like to make some
not really, just a this old hunyuan wsl backup with diffusion pipe that i updated , as for the dataset prep, i can be done with just python scripts, you can get any llm to make them for you, like splitting by scene, converting to 16fps, splitting into 3 sec or 5sec and then another for captioning
I do not like that misty effect that obscures the screen. There is some examples posted here, where it is not present. I am brand new to all this stuff and have found no way to get rid of it. Any advice?
Details
Files
wan22-spitoncamera-36epoc-i2v-low-k3nk.safetensors
Mirrors
151_wan22-spitoncamera-low-k3nk_feb19.safetensors
low.safetensors
SpitOnCamera_LOW.safetensors
wan22-spitoncamera-36epoc-i2v-low-k3nk.safetensors
wan22-spitoncamera-36epoc-i2v-low-k3nk.safetensors
spit-low.safetensors
wan22-spitoncamera-36epoc-i2v-low-k3nk.safetensors
wan22-spitoncamera-36epoc-i2v-low-k3nk.safetensors
wan22-spitoncamera-36epoc-i2v-low-k3nk.safetensors
wan22-spitoncamera-36epoc-i2v-low-k3nk.safetensors
wan22-spitoncamera-36epoc-i2v-low-k3nk.safetensors
spit_lo.safetensors
wan2.2-i2v-low-spitatcam.safetensors
wan22-spitoncamera-36epoc-i2v-low-k3nk.safetensors
wan22-spitoncamera-36epoc-i2v-low-k3nk.safetensors
SpitOnCamera_LOW.safetensors
wan22-spitoncamera-36epoc-i2v-low-k3nk.safetensors
wan22-spitoncamera-36epoc-i2v-low-k3nk.safetensors
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.