{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,6,11]],"date-time":"2025-06-11T05:44:04Z","timestamp":1749620644801},"reference-count":41,"publisher":"Institute of Electronics, Information and Communications Engineers (IEICE)","issue":"12","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEICE Trans. Inf. & Syst."],"published-print":{"date-parts":[[2021,12,1]]},"DOI":"10.1587\/transinf.2021pap0010","type":"journal-article","created":{"date-parts":[[2021,11,30]],"date-time":"2021-11-30T22:45:10Z","timestamp":1638312310000},"page":"2068-2077","source":"Crossref","is-referenced-by-count":1,"title":["A Low-Latency Inference of Randomly Wired Convolutional Neural Networks on an FPGA"],"prefix":"10.1587","volume":"E104.D","author":[{"given":"Ryosuke","family":"KURAMOCHI","sequence":"first","affiliation":[{"name":"Tokyo Institute of Technology"}]},{"given":"Hiroki","family":"NAKAHARA","sequence":"additional","affiliation":[{"name":"Tokyo Institute of Technology"}]}],"member":"532","reference":[{"key":"1","doi-asserted-by":"publisher","unstructured":"[1] V. Badrinarayanan, A. Kendall, and R. Cipolla, \u201cSegnet: A deep convolutional encoder-decoder architecture for image segmentation,\u201d IEEE Trans. Pattern Anal. Mach. Intell., vol.39, no.12, pp.2481-2495, Dec. 2017. 10.1109\/TPAMI.2016.2644615","DOI":"10.1109\/TPAMI.2016.2644615"},{"key":"2","doi-asserted-by":"crossref","unstructured":"[2] W. Liu, D. Anguelov, D. Erhan, C. Szegedy, S.E. Reed, C.Y. Fu, and A.C. Berg, \u201cSSD: Single shot multibox detector,\u201d Proc. European Conf. Comput. Vis. (ECCV), vol.9905, pp.21-37, 2016. 10.1007\/978-3-319-46448-0_2","DOI":"10.1007\/978-3-319-46448-0_2"},{"key":"3","doi-asserted-by":"crossref","unstructured":"[3] J. Redmon and A. Farhadi, \u201cYolo9000: Better, faster, stronger,\u201d 2017 IEEE Conf. Comput. Vis. Pattern Recognit. (CVPR), pp.6517-6525, 2017. 10.1109\/CVPR.2017.690","DOI":"10.1109\/CVPR.2017.690"},{"key":"4","doi-asserted-by":"publisher","unstructured":"[4] S. Ren, K. He, R. Girshick, and J. Sun, \u201cFaster R-CNN: Towards real-time object detection with region proposal networks,\u201d IEEE Trans. Pattern Anal. Mach. Intell., vol.39, no.6, pp.1137-1149, June 2017. 10.1109\/TPAMI.2016.2577031","DOI":"10.1109\/TPAMI.2016.2577031"},{"key":"5","doi-asserted-by":"crossref","unstructured":"[5] Z. Cao, T. Simon, S. Wei, and Y. Sheikh, \u201cRealtime multi-person 2d pose estimation using part affinity fields,\u201d CVPR, pp.1302-1310, 2017.","DOI":"10.1109\/CVPR.2017.143"},{"key":"6","unstructured":"[6] https:\/\/ir.tesla.com\/events\/event-details\/tesla-autonomy-investor-day."},{"key":"7","doi-asserted-by":"crossref","unstructured":"[7] S. Xie, A. Kirillov, R. Girshick, and K. He, \u201cExploring randomly wired neural networks for image recognition,\u201d 2019 IEEE\/CVF Int. Conf. Comput. Vis. (ICCV), pp.1284-1293, 2019. 10.1109\/ICCV.2019.00137","DOI":"10.1109\/ICCV.2019.00137"},{"key":"8","doi-asserted-by":"crossref","unstructured":"[8] R. Kuramochi and H. Nakahara, \u201cAn FPGA-based low-latency accelerator for randomly wired neural networks,\u201d 2020 30th Int. Conf. Field-Programmable Logic and Applications (FPL), pp.298-303, 2020. 10.1109\/FPL50879.2020.00056","DOI":"10.1109\/FPL50879.2020.00056"},{"key":"9","doi-asserted-by":"crossref","unstructured":"[9] O. Russakovsky, J. Deng, H. Su, J. Krause, S. Satheesh, S. Ma, Z. Huang, A. Karpathy, A. Khosla, M. Bernstein, A.C. Berg, and L. Fei-Fei, \u201cImageNet large scale visual recognition challenge,\u201d Int. J. Comput. Vis. (IJCV), vol.115, no.3, pp.211-252, 2015. 10.1007\/s11263-015-0816-y","DOI":"10.1007\/s11263-015-0816-y"},{"key":"10","doi-asserted-by":"publisher","unstructured":"[10] Y. Ma, Y. Cao, S. Vrudhula, and J. Seo, \u201cOptimizing the convolution operation to accelerate deep neural networks on FPGA,\u201d IEEE Trans. Very Large Scale Integr. (VLSI) Syst., vol.26, no.7, pp.1354-1367, July 2018. 10.1109\/TVLSI.2018.2815603","DOI":"10.1109\/TVLSI.2018.2815603"},{"key":"11","doi-asserted-by":"crossref","unstructured":"[11] Y. Niu, R. Kannan, A. Srivastava, and V. Prasanna, \u201cReuse kernels or activations?: a flexible dataflow for low-latency spectral CNN acceleration,\u201d Proc. 2020 ACM\/SIGDA Int. Symp. Field-Programmable Gate Arrays (FPGA), pp.266-276, Association for Computing Machinery, Feb. 2020. 10.1145\/3373087.3375302","DOI":"10.1145\/3373087.3375302"},{"key":"12","doi-asserted-by":"crossref","unstructured":"[12] S. Hadjis and K. Olukotun, \u201cTensorflow to cloud FPGAs: Tradeoffs for accelerating deep neural networks,\u201d FPL, pp.360-366, 2019. 10.1109\/FPL.2019.00064","DOI":"10.1109\/FPL.2019.00064"},{"key":"13","doi-asserted-by":"crossref","unstructured":"[13] Y. Umuroglu, N.J. Fraser, G. Gambardella, M. Blott, P. Leong, M. Jahre, and K. Vissers, \u201cFinn: A framework for fast, scalable binarized neural network inference,\u201d FPGA, pp.65-74, Association for Computing Machinery, Feb. 2017. 10.1145\/3020078.3021744","DOI":"10.1145\/3020078.3021744"},{"key":"14","doi-asserted-by":"crossref","unstructured":"[14] P. Colangelo, N. Nasiri, E. Nurvitadhi, A. Mishra, M. Margala, and K. Nealis, \u201cExploration of low numeric precision deep learning inference using intel\u00ae FPGAs,\u201d 2018 IEEE 26th Annual Int. Symp. Field-Programmable Custom Computing Machines (FCCM), pp.73-80, 2018. 10.1109\/FCCM.2018.00020","DOI":"10.1109\/FCCM.2018.00020"},{"key":"15","unstructured":"[15] A. Krizhevsky, I. Sutskever, and G.E. Hinton, \u201cImagenet classification with deep convolutional neural networks,\u201d Advances in Neural Information Processing Systems 25, ed. F. Pereira, C.J.C. Burges, L. Bottou, and K.Q. Weinberger, vol.1, pp.1097-1105, Curran Associates, Inc., Dec. 2012."},{"key":"16","unstructured":"[16] K. Simonyan and A. Zisserman, \u201cVery deep convolutional networks for large-scale image recognition,\u201d CoRR, vol.abs\/1409.1556, 2015."},{"key":"17","doi-asserted-by":"crossref","unstructured":"[17] C. Szegedy, Wei Liu, Yangqing Jia, P. Sermanet, S. Reed, D. Anguelov, D. Erhan, V. Vanhoucke, and A. Rabinovich, \u201cGoing deeper with convolutions,\u201d CVPR, pp.1-9, June 2015. 10.1109\/CVPR.2015.7298594","DOI":"10.1109\/CVPR.2015.7298594"},{"key":"18","doi-asserted-by":"crossref","unstructured":"[18] K. He, X. Zhang, S. Ren, and J. Sun, \u201cDeep residual learning for image recognition,\u201d CVPR, pp.770-778, June 2016. 10.1109\/CVPR.2016.90","DOI":"10.1109\/CVPR.2016.90"},{"key":"19","unstructured":"[19] A.G. Howard, M. Zhu, B. Chen, D. Kalenichenko, W. Wang, T. Weyand, M. Andreetto, and H. Adam, \u201cMobilenets: Efficient convolutional neural networks for mobile vision applications,\u201d CoRR, vol.abs\/1704.04861, 2017."},{"key":"20","doi-asserted-by":"crossref","unstructured":"[20] M. Sandler, A. Howard, M. Zhu, A. Zhmoginov, and L. Chen, \u201cMobilenetv2: Inverted residuals and linear bottlenecks,\u201d CVPR, pp.4510-4520, 2018.","DOI":"10.1109\/CVPR.2018.00474"},{"key":"21","doi-asserted-by":"crossref","unstructured":"[21] X. Zhang, X. Zhou, M. Lin, and J. Sun, \u201cShufflenet: An extremely efficient convolutional neural network for mobile devices,\u201d CVPR, pp.6848-6856, 2018. 10.1109\/CVPR.2018.00716","DOI":"10.1109\/CVPR.2018.00716"},{"key":"22","doi-asserted-by":"crossref","unstructured":"[22] J. Su, J. Faraone, J. Liu, Y. Zhao, D.B. Thomas, P.H.W. Leong, and P.Y.K. Cheung, \u201cRedundancy-reduced mobilenet acceleration on reconfigurable logic for imagenet classification,\u201d Applied Reconfigurable Computing. Architectures, Tools, and Applications, ed. N. Voros, M. Huebner, G. Keramidas, D. Goehringer, C. Antonopoulos, and P.C. Diniz, vol.10824, pp.16-28, Springer International Publishing, 2018. 10.1007\/978-3-319-78890-6_2","DOI":"10.1007\/978-3-319-78890-6_2"},{"key":"23","doi-asserted-by":"crossref","unstructured":"[23] R. Zhao, X. Niu, and W. Luk, \u201cAutomatic optimising CNN with depthwise separable convolution on FPGA: (abstact only),\u201d FPGA, p.285, Association for Computing Machinery, Feb. 2018. 10.1145\/3174243.3174959","DOI":"10.1145\/3174243.3174959"},{"key":"24","doi-asserted-by":"publisher","unstructured":"[24] B. Liu, D. Zou, L. Feng, S. Feng, P. Fu, and J. Li, \u201cAn FPGA-based CNN accelerator integrating depthwise separable convolution,\u201d Electronics, vol.8, no.3, p.281, March 2019. 10.3390\/electronics8030281","DOI":"10.3390\/electronics8030281"},{"key":"25","doi-asserted-by":"crossref","unstructured":"[25] F. Chollet, \u201cXception: Deep learning with depthwise separable convolutions,\u201d CVPR, pp.1800-1807, 2017. 10.1109\/CVPR.2017.195","DOI":"10.1109\/CVPR.2017.195"},{"key":"26","doi-asserted-by":"publisher","unstructured":"[26] L. Bai, Y. Zhao, and X. Huang, \u201cA CNN accelerator on FPGA using depthwise separable convolution,\u201d IEEE Trans. Circuits Syst. II, Exp. Briefs, vol.65, no.10, pp.1415-1419, Oct. 2018. 10.1109\/TCSII.2018.2865896","DOI":"10.1109\/TCSII.2018.2865896"},{"key":"27","doi-asserted-by":"crossref","unstructured":"[27] D. Wu, Y. Zhang, X. Jia, L. Tian, T. Li, L. Sui, D. Xie, and Y. Shan, \u201cA high-performance CNN processor based on FPGA for mobilenets,\u201d FPL, pp.136-143, 2019. 10.1109\/FPL.2019.00030","DOI":"10.1109\/FPL.2019.00030"},{"key":"28","doi-asserted-by":"crossref","unstructured":"[28] J. Knapheide, B. Stabernack, and M. Kuhnke, \u201cA high throughput mobilenetv2 FPGA implementation based on a flexible architecture for depthwise separable convolution,\u201d FPL, pp.277-283, 2020. 10.1109\/FPL50879.2020.00053","DOI":"10.1109\/FPL50879.2020.00053"},{"key":"29","doi-asserted-by":"crossref","unstructured":"[29] Y. Yang, Q. Huang, B. Wu, T. Zhang, L. Ma, G. Gambardella, M. Blott, L. Lavagno, K. Vissers, J. Wawrzynek, and K. Keutzer, \u201cSynetgy: Algorithm-hardware co-design for convnet accelerators on embedded FPGAs,\u201d FPGA, p.23-32, Association for Computing Machinery, Feb. 2019. 10.1145\/3289602.3293902","DOI":"10.1145\/3289602.3293902"},{"key":"30","unstructured":"[30] S. Ioffe and C. Szegedy, \u201cBatch normalization: Accelerating deep network training by reducing internal covariate shift,\u201d Proc. 32nd Int. Conf. Machine Learning, ed. F. Bach and D. Blei, Proc. Machine Learning Research, vol.37, pp.448-456, 07-09 July 2015."},{"key":"31","unstructured":"[31] K. He, X. Zhang, S. Ren, and J. Sun, \u201cIdentity mappings in deep residual networks,\u201d ArXiv, vol.abs\/1603.05027, 2016."},{"key":"32","unstructured":"[32] M. Lin, Q. Chen, and S. Yan, \u201cNetwork in network,\u201d CoRR, vol.abs\/1312.4400, 2014."},{"key":"33","doi-asserted-by":"crossref","unstructured":"[33] E.N. Gilbert, \u201cRandom graphs,\u201d Ann. Math. Statist., vol.30, no.4, pp.1141-1144, Dec. 1959. 10.1214\/aoms\/1177706098","DOI":"10.1214\/aoms\/1177706098"},{"key":"34","unstructured":"[34] P. Erd\u00f6s and A. R\u00e9nyi, \u201cOn the evolution of random graphs,\u201d Publ. Math. Inst. Hungary. Acad. Sci., vol.5, pp.17-61, 1960. 10.1515\/9781400841356.38"},{"key":"35","doi-asserted-by":"crossref","unstructured":"[35] C. Zhang, P. Li, G. Sun, Y. Guan, B. Xiao, and J. Cong, \u201cOptimizing FPGA-based accelerator design for deep convolutional neural networks,\u201d FPGA, pp.161-170, Feb. 2015. 10.1145\/2684746.2689060","DOI":"10.1145\/2684746.2689060"},{"key":"36","doi-asserted-by":"publisher","unstructured":"[36] D.J.A. Welsh and M.B. Powell, \u201cAn upper bound for the chromatic number of a graph and its application to timetabling problems,\u201d The Computer Journal, vol.10, no.1, pp.85-86, Jan. 1967. 10.1093\/comjnl\/10.1.85","DOI":"10.1093\/comjnl\/10.1.85"},{"key":"37","unstructured":"[37] A. Paszke, S. Gross, F. Massa, A. Lerer, J. Bradbury, G. Chanan, T. Killeen, Z. Lin, N. Gimelshein, L. Antiga, A. Desmaison, A. K\u00f6pf, E. Yang, Z. DeVito, M. Raison, A. Tejani, S. Chilamkurthy, B. Steiner, L. Fang, J. Bai, and S. Chintala, \u201cPytorch: An imperative style, high-performance deep learning library,\u201d Advances in Neural Information Processing Systems 32, ed. H. Wallach, H. Larochelle, A. Beygelzimer, F. d'Alch\u00e9-Buc, E. Fox, and R. Garnett, pp.8024-8035, Curran Associates, Inc., 2019."},{"key":"38","unstructured":"[38] Y. Fu, E. Wu, V. Santhaseelan, K. Denolf, K. Khan, and V. Kathail, \u201cEmbedded vision with int8 optimization on Xilinx devices,\u201d Xilinx, San Jose, CA, USA, White Paper WP490 (v1.0.1), 2017."},{"key":"39","doi-asserted-by":"crossref","unstructured":"[39] B. Jacob, S. Kligys, B. Chen, M. Zhu, M. Tang, A. Howard, H. Adam, and D. Kalenichenko, \u201cQuantization and training of neural networks for efficient integer-arithmetic-only inference,\u201d CVPR, pp.2704-2713, 2018. 10.1109\/CVPR.2018.00286","DOI":"10.1109\/CVPR.2018.00286"},{"key":"40","unstructured":"[40] H. Kung and C. Leiserson, \u201cSystolic arrays (for VLSI),\u201d in Sparse Matrix Proceedings, vol.1, pp.256-282, 1979."},{"key":"41","doi-asserted-by":"crossref","unstructured":"[41] X. Wei, C.H. Yu, P. Zhang, Y. Chen, Y. Wang, H. Hu, Y. Liang, and J. Cong, \u201cAutomated systolic array architecture synthesis for high throughput CNN inference on FPGAs,\u201d 2017 54th ACM\/EDAC\/IEEE Design Automation Conference (DAC), pp.1-6, June 2017. 10.1145\/3061639.3062207","DOI":"10.1145\/3061639.3062207"}],"container-title":["IEICE Transactions on Information and Systems"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.jstage.jst.go.jp\/article\/transinf\/E104.D\/12\/E104.D_2021PAP0010\/_pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2021,12,4]],"date-time":"2021-12-04T03:50:11Z","timestamp":1638589811000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.jstage.jst.go.jp\/article\/transinf\/E104.D\/12\/E104.D_2021PAP0010\/_article"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021,12,1]]},"references-count":41,"journal-issue":{"issue":"12","published-print":{"date-parts":[[2021]]}},"URL":"https:\/\/doi.org\/10.1587\/transinf.2021pap0010","relation":{},"ISSN":["0916-8532","1745-1361"],"issn-type":[{"value":"0916-8532","type":"print"},{"value":"1745-1361","type":"electronic"}],"subject":[],"published":{"date-parts":[[2021,12,1]]},"article-number":"2021PAP0010"}}