Data

Tools

Indexes

Applications

Experiments

Open Source Science

https://github.com/autodistill/autodistill-grounded-sam-2

Use Segment Anything 2, grounded with Florence-2, to auto-label data for use in training vision models.

https://github.com/autodistill/autodistill-grounded-sam-2

Science Score: 13.0%

This score indicates how likely this project is to be science-related based on various indicators:

  • ○
    CITATION.cff file
  • ✓
    codemeta.json file
    Found codemeta.json file
  • ○
    .zenodo.json file
  • ○
    DOI references
  • ○
    Academic publication links
  • ○
    Committers with academic emails
  • ○
    Institutional organization owner
  • ○
    JOSS paper metadata
  • ○
    Scientific vocabulary similarity
    Low similarity (13.1%) to scientific vocabulary

Keywords

autodistill florence-2 grounded-sam segment-anything-2
Last synced: 9 months ago · JSON representation

Repository

Use Segment Anything 2, grounded with Florence-2, to auto-label data for use in training vision models.

Basic Info
  • Host: GitHub
  • Owner: autodistill
  • License: apache-2.0
  • Language: Python
  • Default Branch: main
  • Homepage: https://docs.autodistill.com
  • Size: 30.3 KB
Statistics
  • Stars: 127
  • Watchers: 4
  • Forks: 19
  • Open Issues: 11
  • Releases: 0
Topics
autodistill florence-2 grounded-sam segment-anything-2
Created almost 2 years ago · Last pushed almost 2 years ago
Metadata Files
Readme License

README.md

Autodistill: Grounded SAM 2 Base Model

This repository contains the code implementing Grounded SAM 2 using Florence-2 as a grounding model and Segment Anything 2 as a segmentation model for use with autodistill.

Florence-2 is a zero-shot multimodal model. You can use Florence-2 for open vocabulary object detection. This project uses the object detection capabilities in Florence-2 to ground the SAM 2 model.

Read the full Autodistill documentation.

Read the Grounded SAM 2 Autodistill documentation.

Installation

To use the GroundedSAM Base Model, simply install it along with a Target Model supporting the detection task:

bash pip3 install autodistill-grounded-sam-2 autodistill-yolov8

You can find a full list of detection Target Models on the main autodistill repo.

Quickstart

```python from autodistillgroundedsam_2 import GroundedSAM2 from autodistill.detection import CaptionOntology from autodistill.utils import plot import cv2

define an ontology to map class names to our Grounded SAM 2 prompt

the ontology dictionary has the format {caption: class}

where caption is the prompt sent to the base model, and class is the label that will

be saved for that caption in the generated annotations

then, load the model

base_model = GroundedSAM2( ontology=CaptionOntology( { "person": "person", "shipping container": "shipping container", } ) )

run inference on a single image

results = base_model.predict("logistics.jpeg")

plot( image=cv2.imread("logistics.jpeg"), classes=base_model.ontology.classes(), detections=results )

label all images in a folder called context_images

basemodel.label("./contextimages", extension=".jpeg") ```

License

The code in this repository is licensed under an Apache 2.0 license.

🏆 Contributing

We love your input! Please see the core Autodistill contributing guide to get started. Thank you 🙏 to all our contributors!

Owner

  • Name: Autodistill
  • Login: autodistill
  • Kind: organization
  • Email: autodistill@roboflow.com

Use bigger slower models to train smaller faster ones

GitHub Events

Total
  • Issues event: 5
  • Watch event: 36
  • Issue comment event: 6
  • Pull request event: 2
  • Fork event: 9
Last Year
  • Issues event: 5
  • Watch event: 36
  • Issue comment event: 6
  • Pull request event: 2
  • Fork event: 9

Committers

Last synced: about 1 year ago

All Time
  • Total Commits: 22
  • Total Committers: 1
  • Avg Commits per committer: 22.0
  • Development Distribution Score (DDS): 0.0
Past Year
  • Commits: 22
  • Committers: 1
  • Avg Commits per committer: 22.0
  • Development Distribution Score (DDS): 0.0
Top Committers
Name Email Commits
James j****g@j****g 22
Committer Domains (Top 20 + Academic)
jamesg.blog: 1

Issues and Pull Requests

Last synced: 9 months ago

All Time
  • Total issues: 11
  • Total pull requests: 6
  • Average time to close issues: about 3 hours
  • Average time to close pull requests: N/A
  • Total issue authors: 11
  • Total pull request authors: 3
  • Average comments per issue: 0.36
  • Average comments per pull request: 0.5
  • Merged pull requests: 0
  • Bot issues: 0
  • Bot pull requests: 0
Past Year
  • Issues: 4
  • Pull requests: 6
  • Average time to close issues: 4 minutes
  • Average time to close pull requests: N/A
  • Issue authors: 4
  • Pull request authors: 3
  • Average comments per issue: 0.0
  • Average comments per pull request: 0.5
  • Merged pull requests: 0
  • Bot issues: 0
  • Bot pull requests: 0
View more stats
Top Authors
Issue Authors
  • fikisipi (1)
  • walternat1ve (1)
  • HasnatHridoy (1)
  • mvoodarla (1)
  • casabelle42 (1)
  • Natty02 (1)
  • noorkhokhar99 (1)
  • Parfyonator (1)
  • nikste (1)
  • Bai426 (1)
  • Mechazo11 (1)
Pull Request Authors
  • lukaboljevic (2)
  • AshAnand34 (2)
  • lab176344 (2)
Top Labels
Issue Labels
Pull Request Labels

Packages

  • Total packages: 1
  • Total downloads:
    • pypi 593 last-month
  • Total dependent packages: 0
  • Total dependent repositories: 0
  • Total versions: 1
  • Total maintainers: 1
pypi.org: autodistill-grounded-sam-2

Use Segment Anything 2, grounded with Florence-2, to auto-label data for use in training vision models.

  • Versions: 1
  • Dependent Packages: 0
  • Dependent Repositories: 0
  • Downloads: 593 Last month
Rankings
Dependent packages count: 10.6%
Average: 35.0%
Dependent repos count: 59.5%
Maintainers (1)
capjamesg
Last synced: 10 months ago

Dependencies

requirements.txt pypi
  • autodistill *
  • autodistill_florence_2 *
  • numpy >=1.20.0
  • opencv-python >=4.6.0
  • roboflow *
  • supervision *
  • torch *
setup.py pypi