ESPNet: Efficient Spatial Pyramid of Dilated Convolutions for Semantic Segmentation

Sachin Mehta, Mohammad Rastegari, Anat Caspi, Linda Shapiro, Hannaneh Hajishirzi
3/19/2018
cs.CV

Official code: sacmehta/ESPNet (542 stars, MIT, last commit 2023-06-30).

Abstract

We introduce a fast and efficient convolutional neural network, ESPNet, for semantic segmentation of high resolution images under resource constraints. ESPNet is based on a new convolutional module, efficient spatial pyramid (ESP), which is efficient in terms of computation, memory, and power. ESPNet is 22 times faster (on a standard GPU) and 180 times smaller than the state-of-the-art semantic segmentation network PSPNet, while its category-wise accuracy is only 8% less. We evaluated ESPNet on a variety of semantic segmentation datasets including Cityscapes, PASCAL VOC, and a breast biopsy whole slide image dataset. Under the same constraints on memory and computation, ESPNet outperforms all the current efficient CNN networks such as MobileNet, ShuffleNet, and ENet on both standard metrics and our newly introduced performance metrics that measure efficiency on edge devices. Our network can process high resolution images at a rate of 112 and 9 frames per second on a standard GPU and edge device, respectively.

View on arXivView PDF

Code Implementations(1)

sacmehta/ESPNetOfficial100%
MIT

ESPNet: Efficient Spatial Pyramid of Dilated Convolutions for Semantic Segmentation

542111PythonFeb 18, 20183 years agoMIT
convolutional-neural-networksedge-devicesreal-timesemantic-segmentation

Similar Papers

Efficient Dense Modules of Asymmetric Convolution for Real-Time Semantic Segmentation

88% match1 repo332018

HoughToRadon Transform: New Neural Network Layer for Features Improvement in Projection Space

87% match2024

Fast Neural Architecture Search of Compact Semantic Segmentation Models via Auxiliary Cells

87% match1 repo1512018

Training of Convolutional Networks on Multiple Heterogeneous Datasets for Street Scene Semantic Segmentation

86% match2018

SHARP-Net: A Refined Pyramid Network for Deficiency Segmentation in Culverts and Sewer Pipes

86% match2024

UNet++: Redesigning Skip Connections to Exploit Multiscale Features in Image Segmentation

86% match1 repo2.7k2019

Cite this paper

@article{mehta2018espnet,
  title  = {ESPNet: Efficient Spatial Pyramid of Dilated Convolutions for Semantic Segmentation},
  author = {Sachin Mehta and Mohammad Rastegari and Anat Caspi and Linda Shapiro and Hannaneh Hajishirzi},
  year   = {2018},
  eprint = {1803.06815},
  archivePrefix = {arXiv},
  url    = {https://arxiv.org/abs/1803.06815}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/1803.06815.svg)](https://codeofpaper.com/paper/1803.06815)

Discussion