SAHI: Slicing Aided Hyper Inference

A lightweight vision library for performing large scale object detection & instance segmentation

Object detection and instance segmentation are by far the most important fields of applications in Computer Vision. However, detection of small objects and inference on large images are still major issues in practical usage. Here comes the SAHI to help developers overcome these real-world problems with many vision utilities.

Check this link for a list of competitions that SAHI made us win 🚀

• Introduction to SAHI

• Official paper (NEW)

• Kaggle notebook (NEW)

• Satellite object detection (NEW)

• Error analysis plots & evaluation (NEW)

• Interactive result visualization and inspection (NEW)

• COCO dataset conversion

• Slicing operation notebook

• YOLOX + SAHI demo: (RECOMMENDED)

• YOLOv5 + SAHI walkthrough:

• MMDetection + SAHI walkthrough:

• Detectron2 + SAHI walkthrough: (NEW)

pip install sahi

conda install -c conda-forge shapely

conda install pytorch=1.10.2 torchvision=0.11.3 cudatoolkit=11.3 -c pytorch

pip install yolov5

pip install mmcv-full==1.4.4 -f https://download.openmmlab.com/mmcv/dist/cu113/torch1.10.0/index.html

pip install mmdet==2.21.0

pip install detectron2 -f https://dl.fbaipublicfiles.com/detectron2/wheels/cu113/torch1.10/index.html

Find detailed info on sahi predict command at cli.md.

Find detailed info on image/dataset slicing utilities at slicing.md.

Find detailed info at Error Analysis Plots & Evaluation.

Find detailed info at Interactive Result Visualization and Inspection.

Find detailed info on COCO utilities (yolov5 conversion, slicing, subsampling, filtering, merging, splitting) at coco.md.

Find detailed info on MOT utilities (ground truth dataset creation, exporting tracker metrics in mot challenge format) at mot.md.

If you use this package in your work, please cite it as:

@misc{akyon2022sahi,
      title={Slicing Aided Hyper Inference and Fine-tuning for Small Object Detection}, 
      author={Fatih Cagatay Akyon and Sinan Onur Altinuc and Alptekin Temizel},
      year={2022},
      eprint={2202.06934},
      archivePrefix={arXiv},
      primaryClass={cs.CV}
}

@software{obss2021sahi,
  author       = {Akyon, Fatih Cagatay and Cengiz, Cemil and Altinuc, Sinan Onur and Cavusoglu, Devrim and Sahin, Kadir and Eryuksel, Ogulcan},
  title        = {{SAHI: A lightweight vision library for performing large scale object detection and instance segmentation}},
  month        = nov,
  year         = 2021,
  publisher    = {Zenodo},
  doi          = {10.5281/zenodo.5718950},
  url          = {https://doi.org/10.5281/zenodo.5718950}
}

sahi library currently supports all YOLOv5 models and MMDetection models. Moreover, it is easy to add new frameworks.

All you need to do is, creating a new class in model.py that implements DetectionModel class. You can take the MMDetection wrapper or YOLOv5 wrapper as a reference.

Before opening a PR:

• Install required development packages:

pip install -U -e .[dev]

• Reformat with black and isort:

black . --config pyproject.toml
isort .