Deep Layer Aggregation

Fisher Yu; Dequan Wang; Evan Shelhamer; Trevor Darrell

doi:10.1109/cvpr.2018.00255

Abstract

Visual recognition requires rich representations that span levels from low to high, scales from small to large, and resolutions from fine to coarse. Even with the depth of features in a convolutional network, a layer in isolation is not enough: compounding and aggregating these representations improves inference of what and where. Architectural efforts are exploring many dimensions for network backbones, designing deeper or wider architectures, but how to best aggregate layers and blocks across a network deserves further attention. Although skip connections have been incorporated to combine layers, these connections have been "shallow" themselves, and only fuse by simple, one-step operations. We augment standard architectures with deeper aggregation to better fuse information across layers. Our deep layer aggregation structures iteratively and hierarchically merge the feature hierarchy to make networks with better accuracy and fewer parameters. Experiments across architectures and tasks show that deep layer aggregation improves recognition and resolution compared to existing branching and merging schemes.

Keywords

Fuse (electrical)Computer scienceMerge (version control)Layer (electronics)Aggregate (composite)Artificial intelligenceInferenceDeep learningNetwork architectureInformation retrievalComputer networkEngineering

Affiliated Institutions

Related Publications

ECAPA-TDNN: Emphasized Channel Attention, Propagation and Aggregation in TDNN Based Speaker Verification

Brecht Desplanques , Jenthe Thienpondt , Kris Demuynck

Current speaker verification techniques rely on a neural network to extract speaker representations. The successful x-vector architecture is a Time Delay Neural Network (TDNN) t...

2020 1214 citations

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

Zongwei Zhou , Md Mahfuzur Rahman Siddiquee , Nima Tajbakhsh +1 more

The state-of-the-art models for medical image segmentation are variants of U-Net and fully convolutional networks (FCN). Despite their success, these models have two limitations...

2019 IEEE Transactions on Medical Imaging 3567 citations

Fully convolutional networks for semantic segmentation

Jonathan Long , Evan Shelhamer , Trevor Darrell

Convolutional networks are powerful visual models that yield hierarchies of features. We show that convolutional networks by themselves, trained end-to-end, pixels-to-pixels, ex...

2015 35498 citations

Understanding the difficulty of training deep feedforward neural networks

Xavier Glorot , Yoshua Bengio

Whereas before 2006 it appears that deep multilayer neural networks were not successfully trained, since then several algorithms have been shown to successfully train them, with...

2010 12630 citations

Fully Convolutional Networks for Semantic Segmentation

Evan Shelhamer , Jonathan Long , Trevor Darrell

Convolutional networks are powerful visual models that yield hierarchies of features. We show that convolutional networks by themselves, trained end-to-end, pixels-to-pixels, im...

2016 IEEE Transactions on Pattern Analysis... 10715 citations

Publication Info

Year: 2018
Type: article
Citations: 1501
Access: Closed

External Links

Download PDF (Free) View on DOI.org arXiv Semantic Scholar

Social Impact

Altmetric

Deep Layer Aggregation

PlumX Metrics

Social media, news, blog, policy document mentions

Citation Metrics

1501

OpenAlex

189

Influential

1143

CrossRef

Cite This

APA Style

                            
                                    Fisher Yu, 
                                
                                    Dequan Wang, 
                                
                                    Evan Shelhamer
                                
                                et al.
                            
                            (2018). 
                            Deep Layer Aggregation. 
                            2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition
                            
                            .
                            https://doi.org/10.1109/cvpr.2018.00255

Identifiers

DOI: 10.1109/cvpr.2018.00255
arXiv: 1707.06484

Data Quality

Data completeness: 79%