Improved Imagery Throughput via Cascaded Uncertainty Pruning on U-Net++


  • Mingshi Li KU Leuven
  • Zifu Wang KU Leuven
  • Matthew Blaschko KU Leuven



UNet , Semantic segmentation, Network pruning, Dynamic inference, Satellite imagery


The extensive use of machine learning inferences in real-life earth observation and remote sensing cases has grown over recent years. Network pruning has been carefully studied in various applications to speed up the machine learning workflow, but mainstream pruning strategies often focus on specific connection significancy rather than the sample difficulty. U-Net++ as a well-versed and capable semantic segmentation deep convolutional neural network architecture, as well as its equivalents, are all facing the challenge of overconfidence, which will create barriers for a robust uncertainty-based pruning strategy to be designed. In the following study, we analyzed the efficiency of deep neural networks and semantic segmentation in satellite imagery analysis, and proposed a new tailored workf low of dynamic pruning for U-Net++ by combining the ideas of network calibration and uncertainty and defining the inference complexity of network input samples. We tested and illustrated the capability of this new workflow and delivered a successful comparative study on its effectiveness on the DeepGlobe satellite imagery road extraction dataset and how it can greatly reduce the computational cost with little performance drop.


D. Blalock, J. J. Gonzalez Ortiz, J. Frankle, and J. Guttag. What is the state of neural network pruning? Proceedings of machine learning and systems, 2:129–146, 2020. doi: 10.48550/arXiv.2003.03033.

I. Demir, K. Koperski, D. Lindenbaum, G. Pang, J. Huang, S. Basu, F. Hughes, D. Tuia, and R. Raskar. Deepglobe 2018: A challenge to parse the earth through satellite images. In Proceedings of the IEEE conference on Computer Vision and Pattern Recognition Workshops, pages 172–181, 2018. doi: 10.1109/cvprw.2018.00031.

J. Frankle and M. Carbin. The lottery ticket hypothesis: Finding sparse, trainable neural networks. arXiv preprint arXiv:1803.03635, 2018. doi: 10.48550/arXiv.1803.03635.

Y. Gal and Z. Ghahramani. Dropout as a bayesian approximation: Representing model uncertainty in deep learning. In international conference on machine learning, pages 10501059. PMLR, 2016. doi: 10.48550/arXiv.1506. 02142.

C. Guo, G. Pleiss, Y. Sun, and K. Q. Weinberger. On calibration of modern neural networks. In International Conference on Machine Learning, pages 1321–1330. PMLR, 2017. doi: 10.48550/arXiv.1706.04599.

K. Han, Y. Wang, Q. Tian, J. Guo, C. Xu, and C. Xu. Ghostnet: More features from cheap operations. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 1580–1589, 2020. doi: 10. 48550/arXiv.1911.11907.

S. Han, H. Mao, and W. J. Dally. Deep compression: Compressing deep neural networks with pruning, trained quantization and huffman coding. arXiv preprint arXiv:1510.00149, 2015. doi: 10.48550/arXiv.1510.00149.

S. Han, J. Pool, J. Tran, and W. Dally. Learning both weights and connections for efficient neural network. Advances in neural information processing systems, 28, 2015. doi: 10.48550/arXiv.1506.02626.

Y. LeCun, J. Denker, and S. Solla. Optimal brain damage. Advances in neural information processing systems, 2, 1989.

N. Lee, T. Ajanthan, and P. H. Torr. Snip: Single-shot network pruning based on connection sensitivity. arXiv preprint arXiv:1810.02340, 2018. doi: 10.48550/arXiv. 1810.02340.

S. Lin, R. Ji, C. Yan, B. Zhang, L. Cao, Q. Ye, F. Huang, and D. Doermann. Towards optimal structured cnn pruning via generative adversarial learning. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 27902799, 2019. doi: 10.48550/arXiv.1903.09291..

T. Lin, S. U. Stich, L. Barba, D. Dmitriev, and M. Jaggi. Dynamic model pruning with feedback. arXiv preprint arXiv:2006.07253, 2020. doi: 10.48550/arXiv.2006.07253.

J. Liu, Z. Xu, R. Shi, R. C. Cheung, and H. K. So. Dynamic sparse training: Find efficient sparse network from scratch with trainable masked layers. arXiv preprint arXiv:2005.06870, 2020. doi: 10.48550/arXiv. 2005.06870.

Z. Liu, M. Sun, T. Zhou, G. Huang, and T. Darrell. Rethinking the value of network pruning. arXiv preprint arXiv:1810.05270, 2018. doi: 10.48550/arXiv.1810.05270.

P. Molchanov, A. Mallya, S. Tyree, I. Frosio, and J. Kautz. Importance estimation for neural network pruning. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 11264–11272, 2019. doi: 10.48550/arXiv.1906.10771.

M. P. Naeini, G. Cooper, and M. Hauskrecht. Obtaining well calibrated probabilities using bayesian binning. In Twenty-Ninth AAAI Conference on Artificial Intelligence, 2015. doi: 10.1609/aaai.v29i1.9602.

A. Niculescu-Mizil and R. Caruana. Predicting good probabilities with supervised learning. In Proceedings of the 22nd international conference on Machine learning, pages 625632, 2005. doi: 10.1145/1102351.1102430.

Nvidia. Deep learning performance documentation. https://docs.nvidia. com/deeplearning/performance/ dl-performance-convolutional/index. html, 2022.

J. Platt et al. Probabilistic outputs for support vector machines and comparisons to regularized likelihood methods. Advances in large margin classifiers, 10(3):61–74, 1999.

A. Zhou, Y. Ma, J. Zhu, J. Liu, Z. Zhang, K. Yuan, W. Sun, and H. Li. Learning n: m fine-grained structured sparse neural networks from scratch. arXiv preprint arXiv:2102.04010, 2021. doi: 10.48550/arXiv. 2102.04010.

Z. Zhou, M. M. Rahman Siddiquee, N. Tajbakhsh, and J. Liang. Unet++: A nested u-net architecture for medical image segmentation. In Deep learning in medical image analysis and multimodal learning for clinical decision support, pages 3–11. Springer, 2018. doi: 10.48550/arXiv.1807.10165."