# xSAM
Segment Anything Model (SAM) variants including:
- CoralSCOP
- RepViT-SAM
- EdgeSAM
- MobileSAM
- SAM (original)
combined in the same API (to make life easier).
For more information on different SAM variants, please see the following:
- [_On Efficient Variants of Segment Anything Model: A Survey_](https://arxiv.org/html/2410.04960v1)
## Installation
The code requires `python>=3.8`, as well as `pytorch>=1.7` and `torchvision>=0.8`.
Please follow the instructions [here](https://pytorch.org/get-started/locally/) to install both PyTorch and TorchVision
dependencies. Installing both PyTorch and TorchVision with CUDA support is strongly recommended.
Install xSAM:
```bash
pip install x-segment-anything
```
## Getting Started
The SAM models can be loaded in the following ways:
```python
from x_segment_anything import sam_model_registry, SamPredictor
model_type = "vit_b_coralscop"
model_type = "repvit"
model_type = "edge_sam"
model_type = "vit_t"
model_type = "vit_b"
model_type = "vit_l"
model_type = "vit_h"
sam_checkpoint = "path_to_checkpoints/model_x_weights.pt"
device = "cuda" if torch.cuda.is_available() else "cpu"
x_sam = sam_model_registry[model_type](checkpoint=sam_checkpoint)
x_sam.to(device=device)
x_sam.eval()
predictor = SamPredictor(x_sam)
predictor.set_image(<your_image>)
masks, _, _ = predictor.predict(<input_prompts>)
```
or generate masks for an entire image:
```python
from x_segment_anything import SamAutomaticMaskGenerator
mask_generator = SamAutomaticMaskGenerator(x_sam)
masks = mask_generator.generate(<your_image>)
```
## Model Checkpoints
For convenience, The following model checkpoints are available in the `sam_model_urls` dictionary and can be downloaded
in python:
```python
import requests
from x_segment_anything.build_sam import sam_model_urls
def download_asset(asset_url, asset_path):
response = requests.get(asset_url)
with open(asset_path, 'wb') as f:
f.write(response.content)
model_path = "vit_b_coral_scop.pt"
model_path = "repvit.pt"
model_path = "edge_sam.pt"
model_path = "edge_sam_3x.pt"
model_path = "vit_t.pt"
model_path = "vit_b.pt"
model_path = "vit_l.pt"
model_path = "vit_h.pt"
model = model_path.split(".")[0]
model_url = sam_model_urls[model]
download_asset(model_url, model_path)
```
### Model Checkpoint URLs:
- [coralscop](https://github.com/Jordan-Pierce/CoralSCOP/releases/download/v0.0.1/vit_b_coralscop.pth)
- [repvit](https://huggingface.co/spaces/jameslahm/repvit-sam/resolve/main/repvit_sam.pt)
- [edge_sam](https://huggingface.co/spaces/chongzhou/EdgeSAM/resolve/main/weights/edge_sam.pth)
- [edge_sam_3x](https://huggingface.co/spaces/chongzhou/EdgeSAM/resolve/main/weights/edge_sam_3x.pth)
- [vit_t](https://huggingface.co/spaces/dhkim2810/MobileSAM/resolve/main/mobile_sam.pt)
- [vit_b](https://dl.fbaipublicfiles.com/segment_anything/sam_vit_b_01ec64.pth)
- [vit_l](https://dl.fbaipublicfiles.com/segment_anything/sam_vit_l_0b3195.pth)
- [vit_h](https://dl.fbaipublicfiles.com/segment_anything/sam_vit_h_4b8939.pth)
### Acknowledgements:
- [CoralSCOP](https://github.com/zhengziqiang/CoralSCOP)
- [RepViT-SAM](https://github.com/THU-MIG/RepViT/tree/main)
- [EdgeSAM](https://github.com/chongzhou96/EdgeSAM)
- [MobileSAM](https://github.com/ChaoningZhang/MobileSAM)
- [SAM](https://github.com/facebookresearch/segment-anything)
---
## Disclaimer
This repository is a scientific product and is not official communication of the National
Oceanic and Atmospheric Administration, or the United States Department of Commerce. All NOAA
GitHub project code is provided on an 'as is' basis and the user assumes responsibility for its
use. Any claims against the Department of Commerce or Department of Commerce bureaus stemming from
the use of this GitHub project will be governed by all applicable Federal law. Any reference to
specific commercial products, processes, or services by service mark, trademark, manufacturer, or
otherwise, does not constitute or imply their endorsement, recommendation or favoring by the
Department of Commerce. The Department of Commerce seal and logo, or the seal and logo of a DOC
bureau, shall not be used in any manner to imply endorsement of any commercial product or activity
by DOC or the United States Government.
## License
Software code created by U.S. Government employees is not subject to copyright in the United States
(17 U.S.C. ยง105). The United States/Department of Commerce reserve all rights to seek and obtain
copyright protection in countries other than the United States for Software authored in its
entirety by the Department of Commerce. To this end, the Department of Commerce hereby grants to
Recipient a royalty-free, nonexclusive license to use, copy, and create derivative works of the
Software outside of the United States.
Raw data
{
"_id": null,
"home_page": "https://github.com/Jordan-Pierce/xSAM",
"name": "x-segment-anything",
"maintainer": null,
"docs_url": null,
"requires_python": ">=3.8",
"maintainer_email": null,
"keywords": null,
"author": "Jordan Pierce",
"author_email": "jordan.pierce@noaa.gov",
"download_url": "https://files.pythonhosted.org/packages/aa/42/59d6024a9dba044b4ade003c8d043e394a46d750c6923c9f1c1093da2dda/x_segment_anything-0.0.7.tar.gz",
"platform": null,
"description": "# xSAM\n\nSegment Anything Model (SAM) variants including:\n- CoralSCOP\n- RepViT-SAM \n- EdgeSAM\n- MobileSAM\n- SAM (original) \n\ncombined in the same API (to make life easier).\n\nFor more information on different SAM variants, please see the following:\n- [_On Efficient Variants of Segment Anything Model: A Survey_](https://arxiv.org/html/2410.04960v1)\n\n## Installation\n\nThe code requires `python>=3.8`, as well as `pytorch>=1.7` and `torchvision>=0.8`. \nPlease follow the instructions [here](https://pytorch.org/get-started/locally/) to install both PyTorch and TorchVision \ndependencies. Installing both PyTorch and TorchVision with CUDA support is strongly recommended.\n\nInstall xSAM:\n\n```bash\npip install x-segment-anything\n```\n\n## Getting Started\nThe SAM models can be loaded in the following ways:\n\n```python\nfrom x_segment_anything import sam_model_registry, SamPredictor\n\nmodel_type = \"vit_b_coralscop\"\nmodel_type = \"repvit\"\nmodel_type = \"edge_sam\"\nmodel_type = \"vit_t\"\nmodel_type = \"vit_b\"\nmodel_type = \"vit_l\"\nmodel_type = \"vit_h\"\n\nsam_checkpoint = \"path_to_checkpoints/model_x_weights.pt\"\n\ndevice = \"cuda\" if torch.cuda.is_available() else \"cpu\"\n\nx_sam = sam_model_registry[model_type](checkpoint=sam_checkpoint)\nx_sam.to(device=device)\nx_sam.eval()\n\npredictor = SamPredictor(x_sam)\npredictor.set_image(<your_image>)\nmasks, _, _ = predictor.predict(<input_prompts>)\n```\n\nor generate masks for an entire image:\n\n```python\nfrom x_segment_anything import SamAutomaticMaskGenerator\n\nmask_generator = SamAutomaticMaskGenerator(x_sam)\nmasks = mask_generator.generate(<your_image>)\n```\n\n## Model Checkpoints\nFor convenience, The following model checkpoints are available in the `sam_model_urls` dictionary and can be downloaded \nin python:\n\n```python\nimport requests\nfrom x_segment_anything.build_sam import sam_model_urls\n\ndef download_asset(asset_url, asset_path):\n response = requests.get(asset_url)\n with open(asset_path, 'wb') as f:\n f.write(response.content)\n\nmodel_path = \"vit_b_coral_scop.pt\"\nmodel_path = \"repvit.pt\"\nmodel_path = \"edge_sam.pt\"\nmodel_path = \"edge_sam_3x.pt\"\nmodel_path = \"vit_t.pt\"\nmodel_path = \"vit_b.pt\"\nmodel_path = \"vit_l.pt\"\nmodel_path = \"vit_h.pt\"\n\nmodel = model_path.split(\".\")[0]\nmodel_url = sam_model_urls[model]\n\ndownload_asset(model_url, model_path)\n\n```\n\n### Model Checkpoint URLs:\n- [coralscop](https://github.com/Jordan-Pierce/CoralSCOP/releases/download/v0.0.1/vit_b_coralscop.pth)\n- [repvit](https://huggingface.co/spaces/jameslahm/repvit-sam/resolve/main/repvit_sam.pt)\n- [edge_sam](https://huggingface.co/spaces/chongzhou/EdgeSAM/resolve/main/weights/edge_sam.pth)\n- [edge_sam_3x](https://huggingface.co/spaces/chongzhou/EdgeSAM/resolve/main/weights/edge_sam_3x.pth)\n- [vit_t](https://huggingface.co/spaces/dhkim2810/MobileSAM/resolve/main/mobile_sam.pt)\n- [vit_b](https://dl.fbaipublicfiles.com/segment_anything/sam_vit_b_01ec64.pth)\n- [vit_l](https://dl.fbaipublicfiles.com/segment_anything/sam_vit_l_0b3195.pth)\n- [vit_h](https://dl.fbaipublicfiles.com/segment_anything/sam_vit_h_4b8939.pth)\n\n### Acknowledgements:\n- [CoralSCOP](https://github.com/zhengziqiang/CoralSCOP)\n- [RepViT-SAM](https://github.com/THU-MIG/RepViT/tree/main)\n- [EdgeSAM](https://github.com/chongzhou96/EdgeSAM)\n- [MobileSAM](https://github.com/ChaoningZhang/MobileSAM)\n- [SAM](https://github.com/facebookresearch/segment-anything)\n\n--- \n## Disclaimer\n\nThis repository is a scientific product and is not official communication of the National \nOceanic and Atmospheric Administration, or the United States Department of Commerce. All NOAA \nGitHub project code is provided on an 'as is' basis and the user assumes responsibility for its \nuse. Any claims against the Department of Commerce or Department of Commerce bureaus stemming from \nthe use of this GitHub project will be governed by all applicable Federal law. Any reference to \nspecific commercial products, processes, or services by service mark, trademark, manufacturer, or \notherwise, does not constitute or imply their endorsement, recommendation or favoring by the \nDepartment of Commerce. The Department of Commerce seal and logo, or the seal and logo of a DOC \nbureau, shall not be used in any manner to imply endorsement of any commercial product or activity \nby DOC or the United States Government.\n\n\n## License \n\nSoftware code created by U.S. Government employees is not subject to copyright in the United States \n(17 U.S.C. \u00a7105). The United States/Department of Commerce reserve all rights to seek and obtain \ncopyright protection in countries other than the United States for Software authored in its \nentirety by the Department of Commerce. To this end, the Department of Commerce hereby grants to \nRecipient a royalty-free, nonexclusive license to use, copy, and create derivative works of the \nSoftware outside of the United States.\n",
"bugtrack_url": null,
"license": null,
"summary": null,
"version": "0.0.7",
"project_urls": {
"Homepage": "https://github.com/Jordan-Pierce/xSAM"
},
"split_keywords": [],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "b9bc9a78afa327cdc3d29b57687a984d8a7e511be8e68b09783bbb386b8f6a45",
"md5": "fd243800198069937b58ac0f5ac7a17d",
"sha256": "2886fc8c3c97e872765b998ec6bf15a7c6ea59d9be24927fd6a347c929d07f42"
},
"downloads": -1,
"filename": "x_segment_anything-0.0.7-py3-none-any.whl",
"has_sig": false,
"md5_digest": "fd243800198069937b58ac0f5ac7a17d",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": ">=3.8",
"size": 50312,
"upload_time": "2024-12-11T14:21:54",
"upload_time_iso_8601": "2024-12-11T14:21:54.872030Z",
"url": "https://files.pythonhosted.org/packages/b9/bc/9a78afa327cdc3d29b57687a984d8a7e511be8e68b09783bbb386b8f6a45/x_segment_anything-0.0.7-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "aa4259d6024a9dba044b4ade003c8d043e394a46d750c6923c9f1c1093da2dda",
"md5": "1cacc6be665c7821b10c65b64dfd8986",
"sha256": "191125a7e919f4e00905760dfa678c16c663b0957b8c7b2d6b4cf44c5729a612"
},
"downloads": -1,
"filename": "x_segment_anything-0.0.7.tar.gz",
"has_sig": false,
"md5_digest": "1cacc6be665c7821b10c65b64dfd8986",
"packagetype": "sdist",
"python_version": "source",
"requires_python": ">=3.8",
"size": 40605,
"upload_time": "2024-12-11T14:21:57",
"upload_time_iso_8601": "2024-12-11T14:21:57.247388Z",
"url": "https://files.pythonhosted.org/packages/aa/42/59d6024a9dba044b4ade003c8d043e394a46d750c6923c9f1c1093da2dda/x_segment_anything-0.0.7.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2024-12-11 14:21:57",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "Jordan-Pierce",
"github_project": "xSAM",
"travis_ci": false,
"coveralls": false,
"github_actions": false,
"lcname": "x-segment-anything"
}
JSON
