Student or educator? Sign up with your school email and get 6 months of Pro free.
CoChat
Sign Up
Cited by 3,134

The DOI matches its Crossref record; 22 of the 22 references it cites were found in a registry or an index.

Selective Kernel Networks

Open access
Li, Xiang, Wang, Wenhai, Hu, Xiaolin, Yang, Jian·2019·2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
DOI 10.1109/cvpr.2019.00060
Open this paper in CoChat — free account

Abstract

In standard Convolutional Neural Networks (CNNs), the receptive fields of artificial neurons in each layer are designed to share the same size. It is well-known in the neuroscience community that the receptive field size of visual cortical neurons are modulated by the stimulus, which has been rarely considered in constructing CNNs. We propose a dynamic selection mechanism in CNNs that allows each neuron to adaptively adjust its receptive field size based on multiple scales of input information. A building block called Selective Kernel (SK) unit is designed, in which multiple branches with different kernel sizes are fused using softmax attention that is guided by the information in these branches. Different attentions on these branches yield different sizes of the effective receptive fields of neurons in the fusion layer. Multiple SK units are stacked to a deep network termed Selective Kernel Networks (SKNets). On the ImageNet and CIFAR benchmarks, we empirically show that SKNet outperforms the existing state-of-the-art architectures with lower model complexity. Detailed analyses show that the neurons in SKNet can capture target objects with different scales, which verifies the capability of neurons for adaptively adjusting their receptive field sizes according to the input. The code and models are available at https://github.com/implus/SKNet.

Overview

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1109/cvpr.2019.00060

Cites

22 of 22 references verified0 citation errors · 0 not checkable

Statistics

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Show all 22 references

Research works printed but not linked: 27

  1. 10.1167/15.9.7 · DOI
  2. 10.1073/pnas.89.17.8366 · DOI · Paper page
  3. 10.1016/0006-8993(78)90937-X · DOI · Paper page
  4. 10.1038/11197 · DOI · Paper page
  5. 10.1109/CVPR.2018.00474 · DOI · Paper page
  6. 10.18653/v1/D15-1044 · DOI · Paper page
  7. mnih, 2014, Recurrent models of visual attention, NIPS
  8. nair, 2010, Rectified linear units improve restricted boltzmann machines, ICML
  9. abdi, 2016, Multi-residual networks. arxiv preprint
  10. jia, 2016, Dynamic filter networks, NIPS
  11. krizhevsky, 2009, Learning multiple layers of features from tiny images
  12. 10.1109/ICCV.2015.172 · DOI · Paper page
  13. larochelle, 2010, Learning to combine foveal glimpses with a third-order boltzmann machine, NIPS
  14. 10.1109/CVPR.2017.75 · DOI · Paper page
  15. zhang, 2018, Image super-resolution using very deep residual channel attention networks, 10.1007/978-3-030-01234-2_18 · DOI · Paper page
  16. zhang, 2017, Residual networks of residual networks: Multilevel residual networks, Transactions on Circuits and Systems for Video Technology
  17. 10.1109/34.730558 · DOI · Paper page
  18. jaderberg, 2015, Spatial transformer networks, NIPS
  19. 10.1109/CVPR.2017.200 · DOI · Paper page
  20. chen, 2018, Person search via a mask-guided two-stream cnn model, 10.1007/978-3-030-01234-2_45 · DOI
  21. 10.1109/32.666826 · doi:10.1109/32.666826
  22. chen, 2017, Dual path networks, NIPS
  23. 10.1109/CVPR.2016.503 · DOI · Paper page
  24. 10.1007/978-3-030-01234-2_1 · DOI · Paper page
  25. 10.1109/FG.2018.00088 · DOI
  26. 10.1109/CVPR.2017.634 · DOI · Paper page
  27. 10.1109/CVPR.2015.7298594 · DOI · Paper page

Cited by

Metadata

Authors
Li, X., Wang, W., Hu, X., & Yang, J.
Venue
2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
Publisher
IEEE
Published
2019-06
Pages
510-519
ISSN
1063-6919 (ISSN-L)
DOI
10.1109/cvpr.2019.00060
OpenAlex
W2922509574
Cited by
3,134
References
22
FWCI
91.76
Type
conference paper
Language
en
Open access
green
Licence

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.