{"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/pydmobilenet-improved-version-of-mobilenets","title":"PydMobileNet: Improved Version of MobileNets with Pyramid Depthwise Separable Convolution","arxiv_id":"1811.07083","date":"2018-11-17","proceeding":null,"authors":["Van-Thanh Hoang","Kang-Hyun Jo"],"abstract":"Convolutional neural networks (CNNs) have shown remarkable performance in\nvarious computer vision tasks in recent years. However, the increasing model\nsize has raised challenges in adopting them in real-time applications as well\nas mobile and embedded vision applications. Many works try to build networks as\nsmall as possible while still have acceptable performance. The state-of-the-art\narchitecture is MobileNets. They use Depthwise Separable Convolution\n(DWConvolution) in place of standard Convolution to reduce the size of\nnetworks. This paper describes an improved version of MobileNet, called Pyramid\nMobile Network. Instead of using just a $3\\times 3$ kernel size for\nDWConvolution like in MobileNet, the proposed network uses a pyramid kernel\nsize to capture more spatial information. The proposed architecture is\nevaluated on two highly competitive object recognition benchmark datasets\n(CIFAR-10, CIFAR-100). The experiments demonstrate that the proposed network\nachieves better performance compared with MobileNet as well as other\nstate-of-the-art networks. Additionally, it is more flexible in fine-tuning the\ntrade-off between accuracy, latency and model size than MobileNets.","url_abs":"http://arxiv.org/abs/1811.07083v1","url_pdf":"http://arxiv.org/pdf/1811.07083v1.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":"pydmobilenet-improved-version-of-mobilenets","repo_url":"https://github.com/thanhhvnqb/pydmobilenet_mxnet","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"mxnet","reach":null}],"tasks":[{"task_slug":"object-recognition","task_name":"Object Recognition"}],"methods":[{"method_slug":"convolution","method_name":"Convolution"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}