Densely Connected Convolutional Networks

Huang, Gao; Liu, Zhuang; Maaten, Laurens van der; Weinberger, Kilian Q.

doi:10.48550/arxiv.1608.06993

preprintarXiv (Cornell University)Aug 25, 2016GREEN OA

Densely Connected Convolutional Networks

GHGao Huang ZLZhuang Liu LVLaurens van der Maaten KQKilian Q. Weinberger

Cornell University · Tsinghua University · +1 more institution

Indexed inarxivdatacite

Abstract

Recent work has shown that convolutional networks can be substantially deeper, more accurate, and efficient to train if they contain shorter connections between layers close to the input and those close to the output. In this paper, we embrace this observation and introduce the Dense Convolutional Network (DenseNet), which connects each layer to every other layer in a feed-forward fashion. Whereas traditional convolutional networks with L layers have L connections - one between each layer and its subsequent layer - our network has L(L+1)/2 direct connections. For each layer, the feature-maps of all preceding layers are used as inputs, and its own feature-maps are used as inputs into all subsequent layers.…

Citation impact

1,907

total citations

FWCI: —
Percentile: —
References: 37

Citations per year

Authors

4

Topics & keywords

Topics

Keywords

Benchmark (surveying)
Feature (linguistics)
Computer science
Layer (electronics)
Convolutional neural network
Computation
Reuse
Code (set theory)

UN Sustainable Development Goals

Industry, innovation and infrastructure

No related works found for this paper.