The DOI matches its Crossref record; 19 of the 19 references it cites were found in a registry or an index.
Dynamic Convolutions: Exploiting Spatial Sparsity for Faster Inference
Open accessAbstract
Modern convolutional neural networks apply the same operations on every pixel in an image. However, not all image regions are equally important. To address this inefficiency, we propose a method to dynamically apply convolutions conditioned on the input image. We introduce a residual block where a small gating branch learns which spatial positions should be evaluated. These discrete gating decisions are trained end-to-end using the Gumbel-Softmax trick, in combination with a sparsity criterion. Our experiments on CIFAR, ImageNet, Food-101 and MPII show that our method has better focus on the region of interest and better accuracy than existing methods, at a lower computational complexity. Moreover, we provide an efficient CUDA implementation of our dynamic convolutions using a gather-scatter approach, achieving a significant improvement in inference speed on MobileNetV2 and ShuffleNetV2. On human pose estimation, a task that is inherently spatially sparse, the processing speed is increased by 60% with no loss in accuracy.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
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
Deep Residual Learning for Image Recognition
He, K., Zhang, X., Ren, S., & Sun, J. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.90 · Cited by 235,474
ImageNet: A large-scale hierarchical image database
Deng, J., Dong, W., Socher, R., Li, L.-J., Li, K., & Fei-Fei, L. (2009) · 2009 IEEE Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2009.5206848 · Cited by 65,333
Squeeze-and-Excitation Networks
Hu, J., Shen, L., & Sun, G. (2018) · 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2018.00745 · Cited by 35,667
MobileNetV2: Inverted Residuals and Linear Bottlenecks
Sandler, M., Howard, A., Zhu, M., Zhmoginov, A., & Chen, L.-C. (2018) · 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2018.00474 · Cited by 28,755
Distilling the Knowledge in a Neural Network
Hinton, G., Vinyals, O., & Dean, J. (2015) · arXiv · doi:10.48550/arxiv.1503.02531 · Cited by 14,204
ShuffleNet: An Extremely Efficient Convolutional Neural Network for Mobile Devices
Zhang, X., Zhou, X., Lin, M., & Sun, J. (2017) · arXiv · doi:10.48550/arxiv.1707.01083 · Cited by 9,982
SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <0.5MB model size
Iandola, F. N., Han, S., Moskewicz, M. W., Ashraf, K., Dally, W. J., & Keutzer, K. (2016) · arXiv · doi:10.48550/arxiv.1602.07360 · Cited by 6,201 · cited 1×
“Attempts to improve this trade-off have focused mostly on designing more efficient architectures [21, 39, 44, 53] or compressing existing ones [17, 19, 28, 33, 38, 50].”
BackgroundStacked Hourglass Networks for Human Pose Estimation
Newell, A., Yang, K., & Deng, J. (2016) · Lecture Notes in Computer Science · doi:10.1007/978-3-319-46484-8_29 · Cited by 5,634 · cited 1×
“SACT [12] is not directly applicable on branched architectures such as stacked hourglass networks [34], as it can only halt execution.”
BackgroundEfficient Processing of Deep Neural Networks: A Tutorial and Survey
Sze, V., Chen, Y.-H., Yang, T.-J., & Emer, J. S. (2017) · Proceedings of the IEEE · doi:10.1109/jproc.2017.2761740 · Cited by 4,215 · cited 1×
“Parallel execution on GPU or FPGA relies on the regularity of standard convolutions to pipeline operations [26, 42] and adding element-wise conditional statements might strongly slow down inference.”
Categorical Reparameterization with Gumbel-Softmax
Jang, E., Gu, S., & Poole, B. (2016) · arXiv · doi:10.48550/arxiv.1611.01144 · Cited by 3,616
MnasNet: Platform-Aware Neural Architecture Search for Mobile
Tan, M., Chen, B., Pang, R., Vasudevan, V., Sandler, M., Howard, A., & Le, Q. V. (2019) · 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2019.00293 · Cited by 3,416
FitNets: Hints for Thin Deep Nets
Romero, A., Ballas, N., Kahou, S. E., Chassang, A., Gatta, C., & Bengio, Y. (2014) · arXiv · doi:10.48550/arxiv.1412.6550 · Cited by 2,859
Estimating or Propagating Gradients Through Stochastic Neurons for Conditional Computation
Bengio, Y., Léonard, N., & Courville, A. (2013) · arXiv · doi:10.48550/arxiv.1308.3432 · Cited by 1,910
BranchyNet: Fast inference via early exiting from deep neural networks
Teerapittayanon, S., McDanel, B., & Kung, H. T. (2016) · 2016 23rd International Conference on Pattern Recognition (ICPR) · doi:10.1109/icpr.2016.7900006 · Cited by 1,326
An Analysis of Deep Neural Network Models for Practical Applications
Canziani, A., Paszke, A., & Culurciello, E. (2016) · arXiv · doi:10.48550/arxiv.1605.07678 · Cited by 969 · cited 1×
“, frames processed per second) and accuracy [9, 24].”
BackgroundLow-rank matrix factorization for Deep Neural Network training with high-dimensional output targets
Sainath, T. N., Kingsbury, B., Sindhwani, V., Arisoy, E., & Ramabhadran, B. (2013) · 2013 IEEE International Conference on Acoustics, Speech and Signal Processing · doi:10.1109/icassp.2013.6638949 · Cited by 593
BlockDrop: Dynamic Inference Paths in Residual Networks
Wu, Z., Nagarajan, T., Kumar, A., Rennie, S., Davis, L. S., Grauman, K., & Feris, R. (2018) · 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2018.00919 · Cited by 465
Adaptive Computation Time for Recurrent Neural Networks
Graves, A. (2016) · arXiv · doi:10.48550/arxiv.1603.08983 · Cited by 335 · cited 3×
“The closest work to ours is probably Spatially Adaptive Computation Time (SACT) [12].”
BackgroundMethodSBNet: Sparse Blocks Network for Fast Inference
Ren, M., Pokrovsky, A., Yang, B., & Urtasun, R. (2018) · 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2018.00908 · Cited by 198
Research works printed but not linked: 36
- molchanov, 2016, Pruning convolutional neural networks for resource efficient inference
- mnih, 2014, Recurrent models of visual attention, Advances in Neural Information Processing Systems (NIPS), 2204
- maddison, 0, The concrete distribution: A continuous relaxation of discrete random variables, Proc of the Int Conf on Learning Representations (ICLR)
- lin, 2017, Runtime neural pruning, Advances in Neural Information Processing Systems (NIPS), 2181
- 10.3850/9783981537079_0819 · DOI
- li, 2016, Pruning filters for efficient convnets
- lecun, 1990, Optimal brain damage, Advances in Neural Information Processing Systems (NIPS), 598
- 10.1109/CVPR.2017.684 · DOI · Paper page
- 10.1109/CVPR.2014.471 · DOI · Paper page
- almahairi, 0, Dynamic capacity networks, Proceedings of the International Conference on Machine Learning (ICML), 2549
- 10.1007/978-3-319-46493-0_39 · DOI · Paper page
- 10.1162/neco.1991.3.1.79 · DOI · Paper page
- kim, 2015, Compression of deep convolutional neural networks for fast and low power mobile applications
- 10.1109/CVPR.2016.435 · DOI · Paper page
- krizhevsky, 2009, Learning multiple layers of features from tiny images Technical report
- 10.1109/CVPR.2016.521 · DOI · Paper page
- 10.1109/CVPR.2018.00716 · DOI · Paper page
- 10.1109/CVPR.2019.00363 · DOI · Paper page
- 10.1016/j.robot.2014.11.005 · DOI
- sharma, 2015, Action recognition using visual attention
- 10.1109/CVPR.2017.194 · DOI · Paper page
- gao, 2018, Dynamic channel pruning Feature boosting and suppression
- graham, 2014, Spatially-sparse convolutional neural networks
- graham, 2017, Sub-manifold sparse convolutional networks
- han, 2015, Deep compression Compressing deep neural networks with pruning trained quantization and huffman coding
- bengio, 2015, Conditional computation in neural networks for faster models
- bejnordi, 2019, Batch-shaped channel gated networks
- 10.1007/978-3-642-39593-2_1 · DOI · Paper page
- bossard, 0, Food-10 1–mining discriminative components with random forests, Proceedings of the European Conference on Computer Vision (ECCV), 446
- bolukbasi, 0, Adaptive neural networks for efficient inference, Proc of the International Conference on Machine Learning (ICML), 70, 527
- wen, 2016, Learning structured sparsity in deep neural networks, Advances in Neural Information Processing Systems (NIPS), 2074
- 10.1007/978-3-030-01246-5_1 · DOI · Paper page
- wang, 0, Skipnet: Learning dynamic routing in convolutional networks, Proceedings of the European Conference on Computer Vision (ECCV), 409
- veit, 2016, Residual networks behave like ensembles of relatively shallow networks, Advances in Neural Information Processing Systems (NIPS), 550
- sindhwani, 2015, Structured transforms for small-footprint deep learning, Advances in Neural Information Processing Systems (NIPS), 3088
- 10.1109/CVPR.2016.308 · DOI · Paper page
Cited by
LightGlue: Local Feature Matching at Light Speed
Lindenberger, P., Sarlin, P.-E., & Pollefeys, M. (2023) · 2023 IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51070.2023.01616 · Cited by 943
Dynamic Neural Networks: A Survey
Han, Y., Huang, G., Song, S., Yang, L., Wang, H., & Wang, Y. (2022) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2021.3117837 · Cited by 848
QueryDet: Cascaded Sparse Query for Accelerating High-Resolution Small Object Detection
Yang, C., Huang, Z., & Wang, N. (2022) · 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr52688.2022.01330 · Cited by 634
“Dynamic Convolution [52] uses a small gating network to predict pixel masks and [60] proposes a stochastic sampling and interpolation network.”
BackgroundA-ViT: Adaptive Tokens for Efficient Vision Transformer
Yin, H., Vahdat, A., Alvarez, J. M., Mallya, A., Kautz, J., & Molchanov, P. (2022) · 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr52688.2022.01054 · Cited by 360
Focal Sparse Convolutional Networks for 3D Object Detection
Chen, Y., Li, Y., Zhang, X., Sun, J., & Jia, J. (2022) · 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr52688.2022.00535 · Cited by 325
Metadata
- Authors
- Verelst, T., & Tuytelaars, T.
- Venue
- 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
- Publisher
- IEEE
- Published
- 2020-06
- Pages
- 2317-2326
- ISSN
- 1063-6919 (ISSN-L)
- OpenAlex
- W2991680559
- Cited by
- 170
- References
- 19
- FWCI
- 7.54
- Type
- conference paper
- Language
- en
- Open access
- green
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.