{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/hetconv-heterogeneous-kernel-based","title":"HetConv: Heterogeneous Kernel-Based Convolutions for Deep CNNs","arxiv_id":"1903.04120","date":"2019-03-11","proceeding":"CVPR 2019 6","authors":["Pravendra Singh","Vinay Kumar Verma","Piyush Rai","Vinay P. Namboodiri"],"abstract":"We present a novel deep learning architecture in which the convolution\noperation leverages heterogeneous kernels. The proposed HetConv (Heterogeneous\nKernel-Based Convolution) reduces the computation (FLOPs) and the number of\nparameters as compared to standard convolution operation while still\nmaintaining representational efficiency. To show the effectiveness of our\nproposed convolution, we present extensive experimental results on the standard\nconvolutional neural network (CNN) architectures such as VGG \\cite{vgg2014very}\nand ResNet \\cite{resnet}. We find that after replacing the standard\nconvolutional filters in these architectures with our proposed HetConv filters,\nwe achieve 3X to 8X FLOPs based improvement in speed while still maintaining\n(and sometimes improving) the accuracy. We also compare our proposed\nconvolutions with group/depth wise convolutions and show that it achieves more\nFLOPs reduction with significantly higher accuracy.","url_abs":"http://arxiv.org/abs/1903.04120v2","url_pdf":"http://arxiv.org/pdf/1903.04120v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"hetconv-heterogeneous-kernel-based","repo_url":"https://github.com/irvinxav/Efficient-HetConv-Heterogeneous-Kernel-Based-Convolutions","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[],"methods":[{"method_slug":"1x1-convolution","method_name":"1x1 Convolution"},{"method_slug":"average-pooling","method_name":"Average Pooling"},{"method_slug":"batch-normalization","method_name":"Batch Normalization"},{"method_slug":"bottleneck-residual-block","method_name":"Bottleneck Residual Block"},{"method_slug":"convolution","method_name":"Convolution"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"dropout","method_name":"Dropout"},{"method_slug":"global-average-pooling","method_name":"Global Average Pooling"},{"method_slug":"kaiming-initialization","method_name":"Kaiming Initialization"},{"method_slug":"max-pooling","method_name":"Max Pooling"},{"method_slug":"relu","method_name":"ReLU"},{"method_slug":"residual-block","method_name":"Residual Block"},{"method_slug":"residual-connection","method_name":"Residual Connection"},{"method_slug":"speed","method_name":"SPEED"},{"method_slug":"softmax","method_name":"Softmax"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=1903.04120","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}