ISAT_with_segment_anything [isat-sam]

An Interactive Semi-Automatic Annotation Tool Based on Segment Anything

⭐️ If you find this project helpful, please give it a Star! ⭐️

[中文] [English]

Focusing on the field of image segmentation, we strive to create the best image segmentation annotation software.

Please refers to our latest Documentation in English or 中文文档

News

In version 1.5.2, ISAT supports visual prompt base on SAM3.

Visual prompt
In version 1.5.0, ISAT supports SAM3 and also supports text prompt.

Text prompt

For single category

For multiple categories
In version 1.4.0, ISAT add a plugin system. You can use a small amount of code to extend the functionality of ISAT.

Here are some official plugin examples:
- ISAT_plugin_auto_annotate : An auto-annotation function based on the YOLO model, implemented with just 240 lines of code.
- ISAT_plugin_mask_export : A mask export function, implemented with just 160 lines of code.
For other versions and the release note, please refer to releases

Install

Create a conda environment(recommended, optional)

# create environment
conda create -n isat_env python=3.8

# activate environment
conda activate isat_env

Install
```
pip install isat-sam
```
Run
```
isat-sam
```

Star History

Please support us with a star—it's like a virtual coffee!

Contributors

		...
yatengLG	Alias-z	...

Citation

@misc{ISAT_with_segment_anything,
  title={{ISAT with Segment Anything: An Interactive Semi-Automatic Annotation Tool}},
  author={Ji, Shuwei and Zhang, Hongyuan},
  url={https://github.com/yatengLG/ISAT_with_segment_anything},
  note={Updated on 2025-02-07},
  year={2024},
  version={1.33}
}

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

isat_sam-1.5.2.tar.gz (2.2 MB view details)

Uploaded Dec 31, 2025 Source

File details

Details for the file isat_sam-1.5.2.tar.gz.

File metadata

Download URL: isat_sam-1.5.2.tar.gz
Upload date: Dec 31, 2025
Size: 2.2 MB
Tags: Source
Uploaded using Trusted Publishing? No
Uploaded via: twine/6.2.0 CPython/3.14.2

File hashes

Hashes for isat_sam-1.5.2.tar.gz
Algorithm	Hash digest
SHA256	`a64da9e8e02416dfa22f44941761edfb76c4f336f0f120f05e9c28d01e24bc54`
MD5	`b383de946349d25063439af8778262cc`
BLAKE2b-256	`d193b727874b07cdea5da3df717382cb07b1c9285a2ab0d8bff90f0fd8c23511`