ESPNet: Efficient Spatial Pyramid of Dilated Convolutions for Semantic Segmentation

@article{Mehta2018ESPNetES,
  title={ESPNet: Efficient Spatial Pyramid of Dilated Convolutions for Semantic Segmentation},
  author={Sachin Mehta and M. Rastegari and Anat Caspi and L. Shapiro and Hannaneh Hajishirzi},
  journal={ArXiv},
  year={2018},
  volume={abs/1803.06815}
}
  • Sachin Mehta, M. Rastegari, +2 authors Hannaneh Hajishirzi
  • Published 2018
  • Computer Science
  • ArXiv
  • We introduce a fast and efficient convolutional neural network, ESPNet, for semantic segmentation of high resolution images under resource constraints. [...] Key Result 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.Expand Abstract
    Fast-SCNN: Fast Semantic Segmentation Network
    • 51
    • Open Access
    ESPNetv2: A Light-Weight, Power Efficient, and General Purpose Convolutional Neural Network
    • 53
    • Open Access
    Efficient Dense Modules of Asymmetric Convolution for Real-Time Semantic Segmentation
    • 23
    • Highly Influenced
    • Open Access
    Searching for MobileNetV3
    • 321
    • Open Access
    Lednet: A Lightweight Encoder-Decoder Network for Real-Time Semantic Segmentation
    • 11
    • Highly Influenced
    • Open Access
    CGNet: A Light-weight Context Guided Network for Semantic Segmentation
    • 30
    • Highly Influenced
    • Open Access
    Efficient Video Semantic Segmentation with Labels Propagation and Refinement
    • 7
    • Open Access
    ELASTIC: Improving CNNs With Dynamic Scaling Policies
    • 13
    • Open Access

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 65 REFERENCES
    U-Net: Convolutional Networks for Biomedical Image Segmentation
    • 15,306
    • Open Access
    Fully Convolutional Networks for Semantic Segmentation
    • 7,145
    • Highly Influential
    • Open Access
    DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs
    • 5,188
    • Highly Influential
    • Open Access
    Very Deep Convolutional Networks for Large-Scale Image Recognition
    • 39,078
    • Open Access
    ImageNet classification with deep convolutional neural networks
    • 51,866
    • Open Access
    Going deeper with convolutions
    • 19,800
    • Highly Influential
    • Open Access
    SegNet: A Deep Convolutional Encoder-Decoder Architecture for Image Segmentation
    • 3,304
    • Highly Influential
    • Open Access
    Deep Residual Learning for Image Recognition
    • 49,822
    • Highly Influential
    • Open Access
    Rethinking the Inception Architecture for Computer Vision
    • 8,504
    • Highly Influential
    • Open Access
    Understanding Convolution for Semantic Segmentation
    • 490
    • Open Access