Speech Bubble Remove and Copy Tool

This is a tool to remove or copy the speech bubbles.

• Remove all speech bubbles from the image. (source image / cleaned image)
• (As you can see from this example, not all speech bubbles can be captured completely. Or maybe the sample images aren't suitable...)

• Copy all speech bubbles in image A to image B. (source image(A) / base image(B) / base image with speech bubble)

Installation(for windows)

Python 3.10 and git client must be installed

git clone https://github.com/s9roll7/speech_bubble_remove_and_copy.git
cd speech_bubble_remove_and_copy
py -3.10 -m venv venv
venv\Scripts\activate.bat
# Please install torch according to your environment.(https://pytorch.org/get-started/locally/)
python -m pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu121
pip install -r requirements.txt

How To Use

run launch_cmd.bat

# remove
python main.py remove SRC_IMAGE_DIR_PATH

#!! Create base images based on the clean images.
#!! and put it in BASE_IMAGE_DIR with the "same file name".

# copy
python main.py copy SRC_IMAGE_DIR_PATH BASE_IMAGE_DIR_PATH

or

# remove
#  prepare your project directory
#  my_proj_1/
#        src/ <--- put source images here


#  with comic panel extraction
python main.py remove_proj PROJ_DIR_PATH

or

#  without comic panel extraction
python main.py remove_proj PROJ_DIR_PATH --split=False

# copy
#  prepare your project directory
#  my_proj_1/
#        src/ <--- put source images here
#        base/ <--- put base images here

#!! Create base images based on the clean images.
#!! and put it in "my_proj_1/base" with the "same file name".

python main.py copy_proj PROJ_DIR_PATH

Advanced Settings

Switching models

If you want to include text as well as speech bubbles in the process, you need to use a different model.
Download the "adetailerForTextSpeech" model from civitai, Place it in the following location.
data/models/adetailerForTextSpeech_v20/unwantedV10x.pt
Use the following command.

# remove
python main.py remove_proj PROJ_DIR_PATH --model_type=1

# copy
python main.py copy_proj PROJ_DIR_PATH --model_type=1

If you want to use a different model, edit YOLO_SEG_MODEL_LOCATION(in bubble_tool.py)

Changelog

2024-12-22

Added comic panel extraction function
Changed lama model

Related resources

• simple-lama-inpainting
• yolov8m_seg-speech-bubble
• magi
• AnimeMangaInpainting