{"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/quantized-convolutional-neural-networks-for","title":"Quantized Convolutional Neural Networks for Mobile Devices","arxiv_id":"1512.06473","date":"2015-12-21","proceeding":"CVPR 2016 6","authors":["Jiaxiang Wu","Cong Leng","Yuhang Wang","Qinghao Hu","Jian Cheng"],"abstract":"Recently, convolutional neural networks (CNN) have demonstrated impressive\nperformance in various computer vision tasks. However, high performance\nhardware is typically indispensable for the application of CNN models due to\nthe high computation complexity, which prohibits their further extensions. In\nthis paper, we propose an efficient framework, namely Quantized CNN, to\nsimultaneously speed-up the computation and reduce the storage and memory\noverhead of CNN models. Both filter kernels in convolutional layers and\nweighting matrices in fully-connected layers are quantized, aiming at\nminimizing the estimation error of each layer's response. Extensive experiments\non the ILSVRC-12 benchmark demonstrate 4~6x speed-up and 15~20x compression\nwith merely one percentage loss of classification accuracy. With our quantized\nCNN model, even mobile devices can accurately classify images within one\nsecond.","url_abs":"http://arxiv.org/abs/1512.06473v3","url_pdf":"http://arxiv.org/pdf/1512.06473v3.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":"quantized-convolutional-neural-networks-for","repo_url":"https://github.com/jiaxiang-wu/quantized-cnn","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":{"status":"ok"}}],"tasks":[{"task_slug":"classification","task_name":"General Classification"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1512.06473","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}