https://github.com/ae9is/ezsam

Extract objects from images or video via text prompt
https://github.com/ae9is/ezsam

ai background extract filter foreground grounded-segment-anything groundingdino image-editing object-detection prompt remove segment segment-anything select subtract video-editing

Last synced: 7 days ago
JSON representation

Extract objects from images or video via text prompt

• Host: GitHub
• URL: https://github.com/ae9is/ezsam
• Owner: ae9is
• License: agpl-3.0
• Created: 2024-01-09T04:52:31.000Z (11 months ago)
• Default Branch: main
• Last Pushed: 2024-08-01T07:38:23.000Z (4 months ago)
• Last Synced: 2024-10-01T15:41:17.589Z (about 2 months ago)
• Topics: ai, background, extract, filter, foreground, grounded-segment-anything, groundingdino, image-editing, object-detection, prompt, remove, segment, segment-anything, select, subtract, video-editing
• Language: Python
• Homepage: https://www.ezsam.org/
• Size: 7.07 MB
• Stars: 5
• Watchers: 3
• Forks: 0
• Open Issues: 2
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
*Update: [SAM 2 extends SAM to video](https://github.com/facebookresearch/sam2). You might still find ezsam useful for small GIFs or image batches.*

# ezsam (easy segment anything model)

A command line and gui tool to segment images and video via text prompts.

Input images and videos, describe the subjects or objects you want to keep, and output new images and videos with the background removed.

**Check out the docs! [ezsam.org](https://www.ezsam.org)**



  

  

  









## Why?

Meta's [Segment Anything](https://github.com/facebookresearch/segment-anything) is a powerful tool for separating parts of images,

but requires coordinate prompts—either bounding boxes or points.

And manual prompt generation is tedious for large collections of still images or video.

In contrast, text-based prompts describing the object(s) in the foreground to segment can be constant.

Inspired by [Grounded-Segment-Anything](https://github.com/IDEA-Research/Grounded-Segment-Anything),

this project tries to package a simpler to use tool.

If you're not interested in text-based prompts with Segment Anything, 

check out [rembg](https://github.com/danielgatis/rembg).

## How does it work?

The foreground is selected using text prompts to [GroundingDINO](https://github.com/IDEA-Research/GroundingDINO) to detect objects.

Image segments are generated using [Segment Anything](https://github.com/facebookresearch/segment-anything) 

or [Segment Anything HQ (SAM-HQ)](https://github.com/SysCV/SAM-HQ).

## Quick start

```bash

# Ubuntu 22.04, Python 3.9 - 3.11

pip install ezsam

sudo apt install ffmpeg imagemagick

ezsam --help

ezsam-gui

```

For more detailed info, see the documentation site here: [ezsam.org](https://www.ezsam.org)