SwiftNet

Implementation of SwiftNet:Real-time Video Object Segmentation.

Requirements

• Python >= 3.6
• Pytorch 1.5
• Numpy
• Pillow
• opencv-python
• scipy
• tqdm

Training

• The training pipeline of Swiftnet is similar with the training pipeline of STM. You could refer to our reproduced STM training code.

Inference

Usage

python eval.py -g 0 -y 17 -s val -D 'path to davis'

Performance

Performance on Davis-17 val set.

backbone	J&F	J	F	FPS	weights
resnet-18	77.6	75.5	79.7	65	link

Note: The fps is tested on one P100, which does not include the time of image loading and evaluation.

Acknowledgement

This codebase borrows the code and structure from official STM repository.

Citation

@inproceedings{wang2021swiftnet,
  title={SwiftNet: Real-time Video Object Segmentation},
  author={Wang, Haochen and Jiang, Xiaolong and Ren, Haibing and Hu, Yao and Bai, Song},
  booktitle={Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition},
  pages={1296--1305},
  year={2021}
}