Skip to content

A graphical annotation tool to address different Computer Vision tasks.

Notifications You must be signed in to change notification settings

OpenSciML/CvStudio

Repository files navigation

CV-Studio

CV-Studio is a graphical annotation tool to address different Computer Vision tasks.

CV-Studio is developed in Python, Qt, SQLite and uses PyTorch's resources to train deep learning models.

Interface

CVStudio supports:

Datasets:

  • Create and manage your datasets for images.
  • Manually annotate images:
    • Using a label system for classification problems.
    • Using a bounding box for localization and object detection problems.
    • Using a polygon tool or freehand selection for segmentation tasks.
  • Auto-annotate images with a pretrained model to continue tagging the images by your own.

Watch a demo video

Roadmap

  • Datasets: Annotations for videos.
  • Platforms: macOS support.

Installation

Note: CV-Studio only have been developed and tested on Windows, and linux. Future platforms are in the roadmap.

1. Install cvstudio

pip install cvstudio

2. Install pytorch

  • Using GPU:
    pip install --pre torch torchvision -f https://download.pytorch.org/whl/nightly/cu101/torch_nightly.html             
  • Using CPU:
   pip install --pre torch torchvision -f https://download.pytorch.org/whl/nightly/cpu/torch_nightly.html        

3. Download pre-trained models

This command must be executed from the CVStudio folder: Windows (PowerShell)

   Invoke-WebRequest -OutFile ./models/MS_DeepLab_resnet_trained_VOC.pth https://data.vision.ee.ethz.ch/csergi/share/DEXTR/MS_DeepLab_resnet_trained_VOC.pth
   Invoke-WebRequest -OutFile ./models/dextr_pascal-sbd.pth https://data.vision.ee.ethz.ch/csergi/share/DEXTR/dextr_pascal-sbd.pth

Linux

    wget https://data.vision.ee.ethz.ch/csergi/share/DEXTR/MS_DeepLab_resnet_trained_VOC.pth -P ./models
    wget https://data.vision.ee.ethz.ch/csergi/share/DEXTR/dextr_pascal-sbd.pth -P ./models

4. Run CVStudio

    cvstudio

Documentation

Check out the wiki.

How to contribute

Send a pull request.

License

Free software: MIT license

Citation: haruiz. CV-Studio. Git code (2019). https://github.com/haruiz/CvStudio

Credits

Images and Icons

Models

Deep Extreme Cut: From Extreme Points to Object Segmentation

About

A graphical annotation tool to address different Computer Vision tasks.

Resources

Stars

Watchers

Forks

Packages

No packages published