Flattening Singular Values of Factorized Convolution for Medical Images
Abstract: Convolutional neural networks (CNNs) have long been the paradigm of choice for robust medical image processing (MIP). Therefore, it is crucial to effectively and efficiently deploy CNNs on devices with different computing capabilities to support computer-aided diagnosis. Many methods employ factorized convolutional layers to alleviate the burden of limited computational resources at the expense of expressiveness. To this end, given weak medical image-driven CNN model optimization, a Singular value equalization generalizer-induced Factorized Convolution (SFConv) is proposed to improve the expressive power of factorized convolutions in MIP models. We first decompose the weight matrix of convolutional filters into two low-rank matrices to achieve model reduction. Then minimize the KL divergence between the two low-rank weight matrices and the uniform distribution, thereby reducing the number of singular value directions with significant variance. Extensive experiments on fundus and OCTA datasets demonstrate that our SFConv yields competitive expressiveness over vanilla convolutions while reducing complexity.
- “Learning both weights and connections for efficient neural network,” Advances in neural information processing systems, vol. 28, 2015.
- “Distilling the knowledge in a neural network,” arXiv preprint arXiv:1503.02531, vol. 2, no. 7, 2015.
- “Accelerating very deep convolutional networks for classification and detection,” IEEE transactions on pattern analysis and machine intelligence, vol. 38, no. 10, pp. 1943–1955, 2015.
- “Factorized convolutional neural networks,” in Proceedings of the IEEE International Conference on Computer Vision Workshops, 2017, pp. 545–553.
- “Compressing convolutional neural networks via factorized convolutional filters,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2019, pp. 3977–3986.
- “Exploiting linear structure within convolutional networks for efficient evaluation,” Advances in neural information processing systems, vol. 27, 2014.
- “Initialization and regularization of factorized neural layers,” arXiv preprint arXiv:2105.01029, 2021.
- “Factorized convolution with spectral normalization for fundus screening,” in 2022 IEEE 19th International Symposium on Biomedical Imaging (ISBI). IEEE, 2022, pp. 1–5.
- “The pascal visual object classes challenge: A retrospective,” International journal of computer vision, vol. 111, no. 1, pp. 98–136, 2015.
- “Vindr-cxr: An open dataset of chest x-rays with radiologist’s annotations,” arXiv preprint arXiv:2012.15029, 2020.
- “Measuring skewness: a forgotten statistic?,” Journal of statistics education, vol. 19, no. 2, 2011.
- “Convolutional neural networks with low-rank regularization,” arXiv preprint arXiv:1511.06067, 2015.
- “Singular value decomposition and principal component analysis,” in A practical approach to microarray data analysis, pp. 91–109. Springer, 2003.
- “Idrid: Diabetic retinopathy–segmentation and grading challenge,” Medical image analysis, vol. 59, pp. 101561, 2020.
- “Rose: a retinal oct-angiography vessel segmentation dataset and new model,” IEEE transactions on medical imaging, vol. 40, no. 3, pp. 928–939, 2020.
- “Deep residual learning for image recognition,” in Proceedings of the IEEE conference on computer vision and pattern recognition, 2016, pp. 770–778.
- “U-net: Convolutional networks for biomedical image segmentation,” in International Conference on Medical image computing and computer-assisted intervention. Springer, 2015, pp. 234–241.
- François Chollet, “Xception: Deep learning with depthwise separable convolutions,” in Proceedings of the IEEE conference on computer vision and pattern recognition, 2017, pp. 1251–1258.
Paper Prompts
Sign up for free to create and run prompts on this paper using GPT-5.
Top Community Prompts
Collections
Sign up for free to add this paper to one or more collections.