Hugging Face's logo

neph1
/
framepack-camera-controls

Diffusers
Model card Files
xet
 Community
How to use from the
Use from the
Diffusers library
pip install -U diffusers transformers accelerate
import torch
from diffusers import DiffusionPipeline

# switch to "mps" for apple devices
pipe = DiffusionPipeline.from_pretrained("neph1/framepack-camera-controls", dtype=torch.bfloat16, device_map="cuda")

prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k"
image = pipe(prompt).images[0]

Loras mentioned in this article: https://huggingface.co/blog/neph1/framepack-camera-control-loras

Trigger word for each is "cam_control". Trigger and training prompts:

  • rotate_360_500.safetensors - "cam_control", "The camera rotates 360 degrees around the subject."
  • hero_zoom_500.safetensors - "hero_zoom", "the camera zooms in on the subjects face."
  • cam_control_rotate_left_600.safetensors - "cam_control", "The camera rotates around the subject.
  • cam_control_zoom_out.safetensors - "cam_control", "The camera zooms out while focusing on the subject."

    If the training video bleeds through (like a grey cube appearing in the generation), lower the strength. Between 0.5 and 1.0

    250605: Added two more loras

    slow_zoom training video slow_zoom result

    Trigger word: slow_zoom.

    handheld_camera training video handheld_camera result

    Trigger word: handheld_camera (camera not guaranteed)
    • Downloads last month
    62
    Inference Providers NEW
    This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

    Model tree for neph1/framepack-camera-controls

    Base model

    lllyasviel/FramePackI2V_HY
    Finetuned
    (3)
    this model