{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,25]],"date-time":"2026-03-25T16:07:48Z","timestamp":1774454868207,"version":"3.50.1"},"reference-count":48,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","license":[{"start":{"date-parts":[[2021,1,1]],"date-time":"2021-01-01T00:00:00Z","timestamp":1609459200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by-nc-nd\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100010418","name":"Institute for Information & communications Technology Planning & Evaluation (IITP) grant funded by the Korean Government","doi-asserted-by":"publisher","award":["2017-0-00142"],"award-info":[{"award-number":["2017-0-00142"]}],"id":[{"id":"10.13039\/501100010418","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Access"],"published-print":{"date-parts":[[2021]]},"DOI":"10.1109\/access.2021.3133100","type":"journal-article","created":{"date-parts":[[2021,12,6]],"date-time":"2021-12-06T20:54:48Z","timestamp":1638824088000},"page":"164245-164255","source":"Crossref","is-referenced-by-count":26,"title":["Performance Evaluation of INT8 Quantized Inference on Mobile GPUs"],"prefix":"10.1109","volume":"9","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-7747-2143","authenticated-orcid":false,"given":"Sumin","family":"Kim","sequence":"first","affiliation":[],"role":[{"role":"author","vocab":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6734-8648","authenticated-orcid":false,"given":"Gunju","family":"Park","sequence":"additional","affiliation":[],"role":[{"role":"author","vocab":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9802-2109","authenticated-orcid":false,"given":"Youngmin","family":"Yi","sequence":"additional","affiliation":[],"role":[{"role":"author","vocab":"crossref"}]}],"member":"263","reference":[{"key":"ref39","first-page":"448","article-title":"Batch normalization: Accelerating deep network training by reducing internal covariate shift","author":"ioffe","year":"2015","journal-title":"Proc Int Conf Mach Learn"},{"key":"ref38","year":"2019","journal-title":"Gemmlowp"},{"key":"ref33","year":"2018","journal-title":"Qnnpack"},{"key":"ref32","first-page":"8026","article-title":"Pytorch: An imperative style, high-performance deep learning library","volume":"32","author":"paszke","year":"2019","journal-title":"Proc Adv Neural Inf Process Syst"},{"key":"ref31","author":"abadi","year":"2015","journal-title":"TensorFlow Large-Scale Machine Learning on Heterogeneous Systems"},{"key":"ref30","year":"2017","journal-title":"TENSOR"},{"key":"ref37","year":"2017","journal-title":"Low Precision Inference on GPU"},{"key":"ref36","article-title":"Analysis and optimization of convolutional neural network architectures","author":"thoma","year":"2017","journal-title":"arXiv 1707 09725"},{"key":"ref35","year":"2019","journal-title":"ARM Integer Dot Product"},{"key":"ref34","year":"2018","journal-title":"Mali-G76 Taking High-End Graphics to the Next Level"},{"key":"ref10","year":"2017","journal-title":"ARM Computer Library"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.14495\/jsiaml.6.81"},{"key":"ref11","article-title":"On-device neural net inference with mobile GPUs","author":"lee","year":"2019","journal-title":"arXiv 1907 01989"},{"key":"ref12","year":"2018","journal-title":"Caffe"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1109\/TCAD.2019.2944584"},{"key":"ref14","first-page":"1025","article-title":"Optimizing CNN model inference on CPUs","author":"liu","year":"2019","journal-title":"Proc USENIX Annu Tech Conf (USENIX ATC)"},{"key":"ref15","article-title":"Scheduling computation graphs of deep learning models on manycore CPUs","author":"tang","year":"2018","journal-title":"arXiv 1807 09667"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1109\/IPDPS.2016.119"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1109\/IPSN.2016.7460664"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1145\/3081333.3081360"},{"key":"ref19","year":"2019","journal-title":"Tensorflow Model Optimization–Quantization"},{"key":"ref28","year":"2020","journal-title":"PyTorch"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1109\/MDAT.2020.2968258"},{"key":"ref27","year":"2018","journal-title":"TensorFlow Lite"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1109\/ICCVW.2019.00447"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.435"},{"key":"ref29","year":"2019","journal-title":"Coreml"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1109\/HPCA.2019.00048"},{"key":"ref8","article-title":"MobileNets: Efficient convolutional neural networks for mobile vision applications","author":"howard","year":"2017","journal-title":"arXiv 1704 04861"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00474"},{"key":"ref2","article-title":"A survey of model compression and acceleration for deep neural networks","author":"cheng","year":"2017","journal-title":"arXiv 1710 09282"},{"key":"ref9","year":"2018","journal-title":"ARM Announces Mali-G76 GPU Scaling up bifrost"},{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-01234-2_48"},{"key":"ref46","year":"2019","journal-title":"Galaxy Note10+"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00286"},{"key":"ref45","year":"2018","journal-title":"The Samsung Galaxy S9 and S9+Review"},{"key":"ref48","year":"2019","journal-title":"ODROID-N2"},{"key":"ref22","article-title":"Quantizing deep convolutional networks for efficient inference: A whitepaper","author":"krishnamoorthi","year":"2018","journal-title":"arXiv 1806 08342"},{"key":"ref47","author":"frumusanu","year":"2019","journal-title":"Galaxy Note10+–Full Phone Specifications"},{"key":"ref21","year":"2017","journal-title":"8-bit inference with tensorrt"},{"key":"ref42","article-title":"On improving the numerical stability of Winograd convolutions","author":"vincent","year":"2017","journal-title":"Proc Int Conf Learn Represent Workshop"},{"key":"ref24","article-title":"Efficient Winograd convolution via integer arithmetic","author":"meng","year":"2019","journal-title":"arXiv 1901 01965"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-05057-3_15"},{"key":"ref23","first-page":"1","article-title":"$\\mu$\nLayer: Low latency on-device inference using cooperative single-layer acceleration and processor-friendly quantization","author":"kim","year":"2019","journal-title":"Proc 14th EuroSys Conf"},{"key":"ref44","author":"frumusanu","year":"2018","journal-title":"The Samsung Galaxy S9 and S9+ Review"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP40776.2020.9054562"},{"key":"ref43","first-page":"1","article-title":"Efficient winograd-based convolution kernel implementation on edge devices","author":"xygkis","year":"2018","journal-title":"Proc 55th Annu Design Autom Conf"},{"key":"ref25","article-title":"INT8 Winograd acceleration for Conv1D equipped ASR models deployed on mobile devices","author":"yao","year":"2020","journal-title":"arXiv 2010 14841"}],"container-title":["IEEE Access"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/6287639\/9312710\/09638444.pdf?arnumber=9638444","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,3,28]],"date-time":"2022-03-28T21:16:40Z","timestamp":1648502200000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/9638444\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021]]},"references-count":48,"URL":"https:\/\/doi.org\/10.1109\/access.2021.3133100","relation":{},"ISSN":["2169-3536"],"issn-type":[{"value":"2169-3536","type":"electronic"}],"subject":[],"published":{"date-parts":[[2021]]}}}