深度学习论文: Lightweight and Progressively-Scalable Networks for Semantic Segmentation及其PyTorch实现

深度学习论文: Lightweight and Progressively-Scalable Networks for Semantic Segmentation及其PyTorch实现
Lightweight and Progressively-Scalable Networks for Semantic Segmentation
PDF: https://arxiv.org/pdf/2207.13600.pdf
PyTorch代码: https://github.com/shanglianlm0525/CvPytorch
PyTorch代码: https://github.com/shanglianlm0525/PyTorch-Networks

1 概述

LPSNet 首先构建一个微型网络。然后，LPSNet 通过一次扩展单个维度（卷积块的数量、通道的数量或输入分辨率）逐步将微型网络扩展到更大的网络，以满足最佳速度/准确性权衡。