ComfyUI Extension: segment anything

Authored by storyicon

Created

Updated

762 stars

Based on GroundingDino and SAM, use semantic strings to segment any element in an image. The comfyui version of sd-webui-segment-anything.

README

ComfyUI Segment Anything

This project is a ComfyUI version of https://github.com/continue-revolution/sd-webui-segment-anything. At present, only the most core functionalities have been implemented. I would like to express my gratitude to continue-revolution for their preceding work on which this is based.

example

I have ensured consistency with sd-webui-segment-anything in terms of output when given the same input.

Requirements

Please ensure that you have installed Python dependencies using the following command:

pip3 install -r requirements.txt

Models

The models will be automatically downloaded when used. You can also manually download them according to the table below. If the automatic download is slow, you can set the HTTP_PROXY and HTTPS_PROXY environment variables to use a proxy.

bert-base-uncased

You can download the model from https://huggingface.co/bert-base-uncased/tree/main into the models/bert-base-uncased folder located in the root directory of ComfyUI, like this:

ComfyUI
    models
        bert-base-uncased
            config.json
            model.safetensors
            tokenizer_config.json
            tokenizer.json
            vocab.txt

You can also skip this step. During the inference process, bert-base-uncased will be automatically downloaded through the transformers library, and its directory is typically ~/.cache/huggingface/hub/models--bert-base-uncased.

GroundingDino

Please directly download the models and configuration files to the models/grounding-dino directory under the ComfyUI root directory, without modifying the file names.

| name | size | config file | model file |
|-|-|-|-| | GroundingDINO_SwinT_OGC | 694MB | download link | download link | | GroundingDINO_SwinB | 938MB | download link | download link |

SAM

Please directly download the model files to the models/sams directory under the ComfyUI root directory, without modifying the file names.

| name | size | model file | |-|-|-|
| sam_vit_h | 2.56GB |download link | | sam_vit_l | 1.25GB |download link | | sam_vit_b | 375MB |download link |
| sam_hq_vit_h | 2.57GB |download link | | sam_hq_vit_l | 1.25GB |download link | | sam_hq_vit_b | 379MB |download link | | mobile_sam | 39MB |download link |

Contribution

Thank you for considering to help out with the source code! Welcome contributions from anyone on the internet, and are grateful for even the smallest of fixes!

If you'd like to contribute to this project, please fork, fix, commit and send a pull request for me to review and merge into the main code base.