LV-UNet: A Lightweight and Vanilla Model for Medical Image Segmentation

TL;DR

LV-UNet is a lightweight, efficient medical image segmentation model that combines pre-trained backbones and re-parametrization for deployment, achieving a good balance of performance and computational efficiency.

Contribution

The paper introduces LV-UNet, a novel lightweight model with a re-parametrization strategy for efficient deployment in medical image segmentation.

Findings

LV-UNet outperforms existing lightweight models on multiple datasets.

The model achieves real-time performance with reduced parameters.

Experimental results show a favorable trade-off between accuracy and computational load.

Abstract

While large models have achieved significant progress in computer vision, challenges such as optimization complexity, the intricacy of transformer architectures, computational constraints, and practical application demands highlight the importance of simpler model designs in medical image segmentation. This need is particularly pronounced in mobile medical devices, which require lightweight, deployable models with real-time performance. However, existing lightweight models often suffer from poor robustness across datasets, limiting their widespread adoption. To address these challenges, this paper introduces LV-UNet, a lightweight and vanilla model that leverages pre-trained MobileNetv3-Large backbones and incorporates fusible modules. LV-UNet employs an enhanced deep training strategy and switches to a deployment mode during inference by re-parametrization, significantly reducing parameter count and computational overhead. Experimental results on ISIC 2016, BUSI, CVC-ClinicDB, CVC-ColonDB, and Kvair-SEG datasets demonstrate a better trade-off between performance and the computational load. The code will be released at https://github.com/juntaoJianggavin/LV-UNet.