{"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/fd-mobilenet-improved-mobilenet-with-a-fast","title":"FD-MobileNet: Improved MobileNet with a Fast Downsampling Strategy","arxiv_id":"1802.03750","date":"2018-02-11","proceeding":null,"authors":["Zheng Qin","Zhaoning Zhang","Xiaotao Chen","Yuxing Peng"],"abstract":"We present Fast-Downsampling MobileNet (FD-MobileNet), an efficient and\naccurate network for very limited computational budgets (e.g., 10-140 MFLOPs).\nOur key idea is applying an aggressive downsampling strategy to MobileNet\nframework. In FD-MobileNet, we perform 32$\\times$ downsampling within 12\nlayers, only half the layers in the original MobileNet. This design brings\nthree advantages: (i) It remarkably reduces the computational cost. (ii) It\nincreases the information capacity and achieves significant performance\nimprovements. (iii) It is engineering-friendly and provides fast actual\ninference speed. Experiments on ILSVRC 2012 and PASCAL VOC 2007 datasets\ndemonstrate that FD-MobileNet consistently outperforms MobileNet and achieves\ncomparable results with ShuffleNet under different computational budgets, for\ninstance, surpassing MobileNet by 5.5% on the ILSVRC 2012 top-1 accuracy and\n3.6% on the VOC 2007 mAP under a complexity of 12 MFLOPs. On an ARM-based\ndevice, FD-MobileNet achieves 1.11$\\times$ inference speedup over MobileNet and\n1.82$\\times$ over ShuffleNet under the same complexity.","url_abs":"http://arxiv.org/abs/1802.03750v1","url_pdf":"http://arxiv.org/pdf/1802.03750v1.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":"fd-mobilenet-improved-mobilenet-with-a-fast","repo_url":"https://github.com/ARM-software/DeepFreeze","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null},{"paper_slug":"fd-mobilenet-improved-mobilenet-with-a-fast","repo_url":"https://github.com/osmr/imgclsmob","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"mxnet","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"fd-mobilenet-improved-mobilenet-with-a-fast","repo_url":"https://github.com/code-implementation1/Code5/tree/main/mobilenetV3_small_x1_0","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":null}],"tasks":[],"methods":[{"method_slug":"average-pooling","method_name":"Average Pooling"},{"method_slug":"batch-normalization","method_name":"Batch Normalization"},{"method_slug":"channel-shuffle","method_name":"Channel Shuffle"},{"method_slug":"convolution","method_name":"Convolution"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"depthwise-convolution","method_name":"Depthwise Convolution"},{"method_slug":"global-average-pooling","method_name":"Global Average Pooling"},{"method_slug":"max-pooling","method_name":"Max Pooling"},{"method_slug":"pointwise-convolution","method_name":"Pointwise Convolution"},{"method_slug":"relu","method_name":"ReLU"},{"method_slug":"residual-connection","method_name":"Residual Connection"},{"method_slug":"shufflenet","method_name":"ShuffleNet"},{"method_slug":"shufflenet-block","method_name":"ShuffleNet Block"},{"method_slug":"softmax","method_name":"Softmax"}],"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}