Overview
This is a motion-control LoRA specifically built for Wan2.2 I2V. It enables a smooth, vertical camera sweep starting from the subject's face and descending down to the feet.
Development Background
While Wan2.2 is powerful, relying solely on text prompts for specific camera paths like "Face-to-Feet" often results in inconsistent movement or lackluster framing. I created this LoRA to provide precise, expressive control that standard prompting can't quite capture.
Exsample Prompt
The camera zooms in on the woman's face and even moves to her feet.
Even if your initial image only shows a woman's face, simply describe the desired outfit in your prompt—Wan2.2 will brilliantly render the full attire as the camera pans down
### [LoRA Usage Policy & Disclaimer]
By downloading or using this LoRA model, you agree to comply with the following terms and conditions:
#### 1. Permitted Use
* Personal Use: This model is intended for personal, educational, and research purposes only.
* Commercial Use: Prohibited
* Redistribution: You may not redistribute or re-upload this model to other platforms without the creator's explicit permission.
#### 2. Prohibited Activities
* Illegal & Harmful Content: You are strictly prohibited from using this model to generate content that involves illegal acts, defamation, or the creation of non-consensual sexual content (Deepfakes).
* Copyright Infringement: You must not use this model to violate the intellectual property rights of any third party.
* Paid Sale: You are not allowed to sell this model or its modified versions on any marketplace.
#### 3. Disclaimer of Liability
* User Responsibility: The user assumes all legal responsibility for any content generated using this model.
* No Warranty: This model is provided "as is" without any warranties. The creator is not responsible for any legal disputes, damages, or losses resulting from the use or misuse of this model.
Description
FAQ
Comments (6)
cool thanks!
inb4 deleted?
This works as intended! Thank you for making it.
Which base model are you using? I'm just getting a blurry mess throughout the camera pan.
I can't get this to work with a horizontal left-to-right pan like you have in your examples. It seems to force a top to bottom pan regardless of whether the person is lying horizontally and the format of the image is landscape. For top-to-bottom pans on a portrait format image, it's definitely an improvement over just prompting alone so it's still a useful lora to have.
"camera pans horizontally along her body" works for me
Details
Files
Wan2.2 I2V Face-to-Feet Motion camera high_noise.safetensors
Mirrors
Wan2.2 I2V Face-to-Feet Motion camera high_noise.safetensors
Wan2.2 I2V Face-to-Feet Motion camera high_noise.safetensors
Wan2.2 I2V Face-to-Feet Motion camera high_noise.safetensors
Wan2.2 I2V Face-to-Feet Motion camera high_noise.safetensors
Wan2.2 I2V Face-to-Feet-high_noise.safetensors
General-Camera-Face-to-Feet Motion_Wan Video 2.2 I2V-A14B_HIGH.safetensors
Wan2.2 I2V Face-to-Feet Motion camera high_noise.safetensors
Wan2.2 I2V Face-to-Feet Motion camera high_noise.safetensors
Wan2.2 I2V Face-to-Feet Motion camera high_noise.safetensors
Wan2.2 I2V Face-to-Feet Motion camera high_noise.safetensors
wan2.2-i2v-high-facetofeet.safetensors
Wan2.2 I2V Face-to-Feet Motion camera high_noise.safetensors
Wan2.2 I2V Face-to-Feet Motion camera high_noise.safetensors
Wan2.2 I2V Face-to-Feet Motion camera high_noise.safetensors
Wan2.2 I2V Face-to-Feet Motion camera high_noise.safetensors
Available On (2 platforms)
Same model published on other platforms. May have additional downloads or version variants.