meta / cutler

Cut and Learn for unsupervised object detection and instance segmentation

  • Public
  • 2.1K runs
  • GitHub
  • Paper
  • License

Run time and cost

This model costs approximately $0.018 to run on Replicate, or 55 runs per $1, but this varies depending on your inputs. It is also open source and you can run it on your own computer with Docker.

This model runs on Nvidia T4 GPU hardware. Predictions typically complete within 80 seconds. The predict time for this model varies significantly based on the inputs.

Readme

Cut and Learn for Unsupervised Object Detection and Instance Segmentation

Cut-and-LEaRn (CutLER) is a simple approach for training object detection and instance segmentation models without human annotations. It outperforms previous SOTA by 2.7 times for AP50 and 2.6 times for AR on 11 benchmarks.

Features

  • We propose MaskCut approach to generate pseudo-masks for multiple objects in an image.
  • CutLER can learn unsupervised object detectors and instance segmentors solely on ImageNet-1K.
  • CutLER exhibits strong robustness to domain shifts when evaluated on 11 different benchmarks across domains like natural images, video frames, paintings, sketches, etc.
  • CutLER can serve as a pretrained model for fully/semi-supervised detection and segmentation tasks.

Citation

@article{wang2023cut,
  title={Cut and Learn for Unsupervised Object Detection and Instance Segmentation},
  author={Wang, Xudong and Girdhar, Rohit and Yu, Stella X and Misra, Ishan},
  journal={arXiv preprint arXiv:2301.11320},
  year={2023}
}