YUSEG: Yolo and Unet is all you need for cell instance segmentation

Bizhe Bai, Jie Tian, Sicong Luo, Tao Wang, Sisuo Lyu
Proceedings of The Cell Segmentation Challenge in Multi-modality High-Resolution Microscopy Images, PMLR 212:1-15, 2023.

Abstract

Cell instance segmentation, which identifies each specific cell area within a mi- croscope image, is helpful for cell analysis. Because of the high computational cost brought on by the large number of objects in the scene, mainstream instance segmentation techniques require much time and computational resources. In this paper, we proposed a two-stage method in which the first stage detects the bounding boxes of cells, and the second stage is segmentation in the detected bounding boxes. This method reduces inference time by more than 30% on images that image size is larger than 1024 pixels by 1024 pixels compared to the mainstream instance segmentation method while maintaining reasonable accuracy without using any external data.

Cite this Paper

BibTeX
@InProceedings{pmlr-v212-bai23a, title = {YUSEG: Yolo and Unet is all you need for cell instance segmentation}, author = {Bai, Bizhe and Tian, Jie and Luo, Sicong and Wang, Tao and Lyu, Sisuo}, booktitle = {Proceedings of The Cell Segmentation Challenge in Multi-modality High-Resolution Microscopy Images}, pages = {1--15}, year = {2023}, editor = {Ma, Jun and Xie, Ronald and Gupta, Anubha and Guilherme de Almeida, José and Bader, Gary D. and Wang, Bo}, volume = {212}, series = {Proceedings of Machine Learning Research}, month = {28 Nov--09 Dec}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v212/bai23a/bai23a.pdf}, url = {https://proceedings.mlr.press/v212/bai23a.html}, abstract = {Cell instance segmentation, which identifies each specific cell area within a mi- croscope image, is helpful for cell analysis. Because of the high computational cost brought on by the large number of objects in the scene, mainstream instance segmentation techniques require much time and computational resources. In this paper, we proposed a two-stage method in which the first stage detects the bounding boxes of cells, and the second stage is segmentation in the detected bounding boxes. This method reduces inference time by more than 30% on images that image size is larger than 1024 pixels by 1024 pixels compared to the mainstream instance segmentation method while maintaining reasonable accuracy without using any external data.} }
Endnote
%0 Conference Paper %T YUSEG: Yolo and Unet is all you need for cell instance segmentation %A Bizhe Bai %A Jie Tian %A Sicong Luo %A Tao Wang %A Sisuo Lyu %B Proceedings of The Cell Segmentation Challenge in Multi-modality High-Resolution Microscopy Images %C Proceedings of Machine Learning Research %D 2023 %E Jun Ma %E Ronald Xie %E Anubha Gupta %E José Guilherme de Almeida %E Gary D. Bader %E Bo Wang %F pmlr-v212-bai23a %I PMLR %P 1--15 %U https://proceedings.mlr.press/v212/bai23a.html %V 212 %X Cell instance segmentation, which identifies each specific cell area within a mi- croscope image, is helpful for cell analysis. Because of the high computational cost brought on by the large number of objects in the scene, mainstream instance segmentation techniques require much time and computational resources. In this paper, we proposed a two-stage method in which the first stage detects the bounding boxes of cells, and the second stage is segmentation in the detected bounding boxes. This method reduces inference time by more than 30% on images that image size is larger than 1024 pixels by 1024 pixels compared to the mainstream instance segmentation method while maintaining reasonable accuracy without using any external data.
APA
Bai, B., Tian, J., Luo, S., Wang, T. & Lyu, S.. (2023). YUSEG: Yolo and Unet is all you need for cell instance segmentation. Proceedings of The Cell Segmentation Challenge in Multi-modality High-Resolution Microscopy Images, in Proceedings of Machine Learning Research 212:1-15 Available from https://proceedings.mlr.press/v212/bai23a.html.

Related Material