svjack/FramePack_Image_Edit_Lora_Early

FramePack Image Edit Early Lora

This repository contains the necessary steps and scripts to generate A edit of the Image using a image-to-video model. The model leverages LoRA (Low-Rank Adaptation) weights and pre-trained components to create Edit Image based on a input Image and textual prompts.

Prerequisites

Before proceeding, ensure that you have the following installed on your system:

• Ubuntu (or a compatible Linux distribution) • Python 3.x • pip (Python package manager) • Git • Git LFS (Git Large File Storage) • FFmpeg

Installation

1. Update and Install Dependencies

   sudo apt-get update && sudo apt-get install cbm git-lfs ffmpeg
   

2. Clone the Repository

   git clone https://huggingface.co/svjack/FramePack_Image_Edit_Lora_Early
   cd FramePack_Image_Edit_Lora_Early
   

3. Install Python Dependencies

   pip install torch torchvision
   pip install -r requirements.txt
   pip install ascii-magic matplotlib tensorboard huggingface_hub datasets
   pip install moviepy==1.0.3
   pip install sageattention==1.0.6
   

4. Download Model Weights

    git clone https://huggingface.co/lllyasviel/FramePackI2V_HY
    git clone https://huggingface.co/hunyuanvideo-community/HunyuanVideo
    git clone https://huggingface.co/Comfy-Org/HunyuanVideo_repackaged
    git clone https://huggingface.co/Comfy-Org/sigclip_vision_384
   

Usage

To Edit a Image, use the fpack_generate_video.py script with the appropriate parameters. Below are examples of how to do it. Make sure use 512x512 as output (this used to train it.)

• 1 Add a cat

• Input

python fpack_generate_video.py \
    --dit FramePackI2V_HY/diffusion_pytorch_model-00001-of-00003.safetensors \
    --vae HunyuanVideo/vae/diffusion_pytorch_model.safetensors \
    --text_encoder1 HunyuanVideo_repackaged/split_files/text_encoders/llava_llama3_fp16.safetensors \
    --text_encoder2 HunyuanVideo_repackaged/split_files/text_encoders/clip_l.safetensors \
    --image_encoder sigclip_vision_384/sigclip_vision_patch14_384.safetensors \
    --image_path xiang_image.jpg \
    --prompt "add a cat into the picture" \
    --video_size 512  512 --fps 30 --infer_steps 25 \
    --attn_mode sdpa --fp8_scaled \
    --vae_chunk_size 32 --vae_spatial_tile_sample_min_size 128 \
    --save_path save --video_sections 1 --output_type latent_images --one_frame_inference zero_post \
    --seed 1234 --lora_multiplier 1.0 --lora_weight framepack_edit_output/framepack-edit-lora-000005.safetensors

• Output

• 2 Change Background

• Input

python fpack_generate_video.py \
    --dit FramePackI2V_HY/diffusion_pytorch_model-00001-of-00003.safetensors \
    --vae HunyuanVideo/vae/diffusion_pytorch_model.safetensors \
    --text_encoder1 HunyuanVideo_repackaged/split_files/text_encoders/llava_llama3_fp16.safetensors \
    --text_encoder2 HunyuanVideo_repackaged/split_files/text_encoders/clip_l.safetensors \
    --image_encoder sigclip_vision_384/sigclip_vision_patch14_384.safetensors \
    --image_path wanye.jpg \
    --prompt "Change the background into a restaurant in anime style. Keep the character's eye colors and white hair unchanged." \
    --video_size 512  512 --fps 30 --infer_steps 25 \
    --attn_mode sdpa --fp8_scaled \
    --vae_chunk_size 32 --vae_spatial_tile_sample_min_size 128 \
    --save_path save --video_sections 1 --output_type latent_images --one_frame_inference zero_post \
    --seed 1234 --lora_multiplier 1.0 --lora_weight framepack_edit_output/framepack-edit-lora-000005.safetensors

• Output

• 3 Place Train into landscape

• Input

python fpack_generate_video.py \
    --dit FramePackI2V_HY/diffusion_pytorch_model-00001-of-00003.safetensors \
    --vae HunyuanVideo/vae/diffusion_pytorch_model.safetensors \
    --text_encoder1 HunyuanVideo_repackaged/split_files/text_encoders/llava_llama3_fp16.safetensors \
    --text_encoder2 HunyuanVideo_repackaged/split_files/text_encoders/clip_l.safetensors \
    --image_encoder sigclip_vision_384/sigclip_vision_patch14_384.safetensors \
    --image_path train.jpg \
    --prompt "place the train into a beautiful landscape" \
    --video_size 512  512 --fps 30 --infer_steps 25 \
    --attn_mode sdpa --fp8_scaled \
    --vae_chunk_size 32 --vae_spatial_tile_sample_min_size 128 \
    --save_path save --video_sections 1 --output_type latent_images --one_frame_inference zero_post \
    --seed 1234 --lora_multiplier 1.0 --lora_weight framepack_edit_output/framepack-edit-lora-000005.safetensors

• Output