{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,10]],"date-time":"2026-10-10T04:30:43Z","timestamp":1791606643979,"version":"4.3.4"},"reference-count":45,"publisher":"Association for Computing Machinery (ACM)","issue":"3","license":[{"start":{"date-parts":[[2018,9,30]],"date-time":"2018-09-30T00:00:00Z","timestamp":1538265600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/www.acm.org\/publications\/policies\/copyright_policy#Background"}],"funder":[{"name":"European Union's HORIZON 2020 Research and Innovation programme","award":["780788 and 732631"],"award-info":[{"award-number":["780788 and 732631"]}]}],"content-domain":{"domain":["dl.acm.org"],"crossmark-restriction":true},"short-container-title":["ACM Trans. Reconfigurable Technol. Syst."],"published-print":{"date-parts":[[2018,9,30]]},"abstract":"<jats:p>\n            Deep convolutional neural networks (CNNs) obtain outstanding results in tasks that require human-level understanding of data, like image or speech recognition. However, their computational load is significant, motivating the development of CNN-specialized accelerators. This work presents NEURA\n            <jats:sc>ghe<\/jats:sc>\n            , a flexible and efficient hardware\/software solution for the acceleration of CNNs on Zynq SoCs. NEURA\n            <jats:sc>ghe<\/jats:sc>\n            leverages the synergistic usage of Zynq ARM cores and of a powerful and flexible Convolution-Specific Processor deployed on the reconfigurable logic. The Convolution-Specific Processor embeds both a convolution engine and a programmable soft core, releasing the ARM processors from most of the supervision duties and allowing the accelerator to be controlled by software at an ultra-fine granularity. This methodology opens the way for cooperative heterogeneous computing: While the accelerator takes care of the bulk of the CNN workload, the ARM cores can seamlessly execute hard-to-accelerate parts of the computational graph, taking advantage of the NEON vector engines to further speed up computation. Through the companion NeuDNN SW stack, NEURA\n            <jats:sc>ghe<\/jats:sc>\n            supports end-to-end CNN-based classification with a peak performance of 169GOps\/s, and an energy efficiency of 17GOps\/W. Thanks to our heterogeneous computing model, our platform improves upon the state-of-the-art, achieving a frame rate of 5.5 frames per second (fps) on the end-to-end execution of VGG-16 and 6.6fps on ResNet-18.\n          <\/jats:p>","DOI":"10.1145\/3284357","type":"journal-article","created":{"date-parts":[[2018,12,12]],"date-time":"2018-12-12T12:49:32Z","timestamp":1544618972000},"page":"1-24","update-policy":"https:\/\/doi.org\/10.1145\/crossmark-policy","source":"Crossref","is-referenced-by-count":55,"title":["NEURA\n            <scp>ghe<\/scp>"],"prefix":"10.1145","volume":"11","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-8106-4641","authenticated-orcid":false,"given":"Paolo","family":"Meloni","sequence":"first","affiliation":[{"name":"Universit\u00e0 di Cagliari, Italy"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Alessandro","family":"Capotondi","sequence":"additional","affiliation":[{"name":"Universit\u00e0 di Bologna, Italy"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Gianfranco","family":"Deriu","sequence":"additional","affiliation":[{"name":"Universit\u00e0 di Cagliari, Italy and T3LAB, Bologna, Italy"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Michele","family":"Brian","sequence":"additional","affiliation":[{"name":"T3LAB, Bologna, Italy"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Francesco","family":"Conti","sequence":"additional","affiliation":[{"name":"Universit\u00e0 di Bologna, Italy and ETH Zurich, Switzerland, Italy"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Davide","family":"Rossi","sequence":"additional","affiliation":[{"name":"Universit\u00e0 di Bologna, Italy"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Luigi","family":"Raffo","sequence":"additional","affiliation":[{"name":"Universit\u00e0 di Cagliari, Italy"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Luca","family":"Benini","sequence":"additional","affiliation":[{"name":"Universit\u00e0 di Bologna, Italy and ETH Zurich, Switzerland, Italy"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"320","published-online":{"date-parts":[[2018,12,12]]},"reference":[{"key":"e_1_2_1_1_1","unstructured":"Altera. 2017. Altera Arria 10. Retrieved from https:\/\/www.altera.com\/products\/fpga\/arria-series\/arria-10\/overview.html.  Altera. 2017. Altera Arria 10. Retrieved from https:\/\/www.altera.com\/products\/fpga\/arria-series\/arria-10\/overview.html."},{"key":"e_1_2_1_2_1","first-page":"1","article-title":"Neurostream: Scalable and energy-efficient deep learning with smart memory cubes","volume":"99","author":"Azarkhish E.","year":"2017","unstructured":"E. Azarkhish , D. Rossi , I. Loi , and L. Benini . 2017 . Neurostream: Scalable and energy-efficient deep learning with smart memory cubes . IEEE Trans. Parallel Distrib. Syst. PP , 99 (2017), 1 -- 1 . E. Azarkhish, D. Rossi, I. Loi, and L. Benini. 2017. Neurostream: Scalable and energy-efficient deep learning with smart memory cubes. IEEE Trans. Parallel Distrib. Syst. PP, 99 (2017), 1--1.","journal-title":"IEEE Trans. Parallel Distrib. Syst. PP"},{"key":"e_1_2_1_3_1","doi-asserted-by":"publisher","DOI":"10.1109\/TETC.2016.2554318"},{"key":"e_1_2_1_4_1","first-page":"1","article-title":"A 803GOp\/s\/W convolutional network accelerator","volume":"99","author":"Cavigelli L.","year":"2016","unstructured":"L. Cavigelli and L. Benini . 2016 . A 803GOp\/s\/W convolutional network accelerator . IEEE Trans. Circ. Syst. Video Technol. 99 (2016), 1 -- 1 . L. Cavigelli and L. Benini. 2016. A 803GOp\/s\/W convolutional network accelerator. IEEE Trans. Circ. Syst. Video Technol. 99 (2016), 1--1.","journal-title":"IEEE Trans. Circ. Syst. Video Technol."},{"key":"e_1_2_1_5_1","doi-asserted-by":"publisher","DOI":"10.1109\/ISCA.2016.40"},{"key":"e_1_2_1_6_1","volume-title":"Proceedings of the Design, Automation 8 Test in Europe Conference 8 Exhibition (DATE\u201915)","author":"Conti F.","unstructured":"F. Conti and L. Benini . 2015. A ultra-low-energy convolution engine for fast brain-inspired vision in multicore clusters . In Proceedings of the Design, Automation 8 Test in Europe Conference 8 Exhibition (DATE\u201915) . EDA Consortium, 683--688. F. Conti and L. Benini. 2015. A ultra-low-energy convolution engine for fast brain-inspired vision in multicore clusters. In Proceedings of the Design, Automation 8 Test in Europe Conference 8 Exhibition (DATE\u201915). EDA Consortium, 683--688."},{"key":"e_1_2_1_7_1","doi-asserted-by":"publisher","DOI":"10.1109\/TCSI.2017.2698019"},{"key":"e_1_2_1_8_1","unstructured":"M. Courbariaux Y. Bengio and J. David. 2015. BinaryConnect: Training deep neural networks with binary weights during propagations. In Advances in Neural Information Processing Systems 28 (NIPS\u201915). Curran Associates Inc. 3123\u20133131.   M. Courbariaux Y. Bengio and J. David. 2015. BinaryConnect: Training deep neural networks with binary weights during propagations. In Advances in Neural Information Processing Systems 28 (NIPS\u201915). Curran Associates Inc. 3123\u20133131."},{"key":"e_1_2_1_9_1","doi-asserted-by":"publisher","DOI":"10.1145\/2749469.2750389"},{"key":"e_1_2_1_10_1","doi-asserted-by":"publisher","DOI":"10.1109\/TVLSI.2017.2654506"},{"key":"e_1_2_1_11_1","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2015.169"},{"key":"e_1_2_1_12_1","volume-title":"Proceedings of the IEEE International Symposium on Circuits and Systems (ISCAS\u201917)","author":"Gokhale V.","unstructured":"V. Gokhale , A. Zaidy , A. X. M. Chang , and E. Culurciello . 2017. Snowflake: An efficient hardware accelerator for convolutional neural networks . In Proceedings of the IEEE International Symposium on Circuits and Systems (ISCAS\u201917) . 1--4. V. Gokhale, A. Zaidy, A. X. M. Chang, and E. Culurciello. 2017. Snowflake: An efficient hardware accelerator for convolutional neural networks. In Proceedings of the IEEE International Symposium on Circuits and Systems (ISCAS\u201917). 1--4."},{"key":"e_1_2_1_13_1","unstructured":"Google. 2017. Build and Train Machine Learning Models on our New Google Cloud TPUs. Retrieved from https:\/\/www.blog.google\/topics\/google-cloud\/google-cloud-offer-tpus-machine-learning\/.  Google. 2017. Build and Train Machine Learning Models on our New Google Cloud TPUs. Retrieved from https:\/\/www.blog.google\/topics\/google-cloud\/google-cloud-offer-tpus-machine-learning\/."},{"key":"e_1_2_1_14_1","unstructured":"A. Hannun et al. 2014. Deep speech: Scaling up end-to-end speech recognition. arXiv:1412.5567.  A. Hannun et al. 2014. Deep speech: Scaling up end-to-end speech recognition. arXiv:1412.5567."},{"key":"e_1_2_1_15_1","unstructured":"K. He X. Zhang S. Ren and J. Sun. 2015. Deep residual learning for image recognition. arxiv:cs.CV\/1512.03385.  K. He X. Zhang S. Ren and J. Sun. 2015. Deep residual learning for image recognition. arxiv:cs.CV\/1512.03385."},{"key":"e_1_2_1_16_1","volume-title":"Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 770--778","author":"He K.","unstructured":"K. He , X. Zhang , S. Ren , and J. Sun . 2016. Deep residual learning for image recognition . In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 770--778 . K. He, X. Zhang, S. Ren, and J. Sun. 2016. Deep residual learning for image recognition. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 770--778."},{"key":"e_1_2_1_17_1","volume-title":"SqueezeNet: AlexNet-level accuracy with 50 fewer parameters and &lt;1MB model size.CoRR abs\/1602.07360","author":"Iandola Forrest N.","year":"2016","unstructured":"Forrest N. Iandola , Matthew W. Moskewicz , Khalid Ashraf , Song Han , William J. Dally , and Kurt Keutzer . 2016. SqueezeNet: AlexNet-level accuracy with 50 fewer parameters and &lt;1MB model size.CoRR abs\/1602.07360 ( 2016 ). Retrieved from http:\/\/dblp.uni-trier.de\/db\/journals\/corr\/corr1602.html#IandolaMAHDK16. Forrest N. Iandola, Matthew W. Moskewicz, Khalid Ashraf, Song Han, William J. Dally, and Kurt Keutzer. 2016. SqueezeNet: AlexNet-level accuracy with 50 fewer parameters and &lt;1MB model size.CoRR abs\/1602.07360 (2016). Retrieved from http:\/\/dblp.uni-trier.de\/db\/journals\/corr\/corr1602.html#IandolaMAHDK16."},{"key":"e_1_2_1_18_1","unstructured":"Image-Net. 2017. Large Scale Visual Recognition Challenge. Retrieved from http:\/\/image-net.org\/.  Image-Net. 2017. Large Scale Visual Recognition Challenge. Retrieved from http:\/\/image-net.org\/."},{"key":"e_1_2_1_19_1","unstructured":"S. Ioffe and C. Szegedy. 2015. Batch normalization: Accelerating deep network training by reducing internal covariate shift. CoRR abs\/1502.03167 (2015). Retrieved from http:\/\/arxiv.org\/abs\/1502.03167.  S. Ioffe and C. Szegedy. 2015. Batch normalization: Accelerating deep network training by reducing internal covariate shift. CoRR abs\/1502.03167 (2015). Retrieved from http:\/\/arxiv.org\/abs\/1502.03167."},{"key":"e_1_2_1_20_1","doi-asserted-by":"publisher","DOI":"10.1145\/3079856.3080246"},{"key":"e_1_2_1_21_1","unstructured":"A. Krizhevsky I. Sutskever and G. Hinton. 2012. Imagenet classification with deep convolutional neural networks. In Advances in Neural Information Processing Systems 25 (NIPS\u201912). Curran Associates Inc. 1097\u20131105.   A. Krizhevsky I. Sutskever and G. Hinton. 2012. Imagenet classification with deep convolutional neural networks. In Advances in Neural Information Processing Systems 25 (NIPS\u201912). Curran Associates Inc. 1097\u20131105."},{"key":"e_1_2_1_22_1","volume-title":"Proceedings of the 27th International Conference on Field Programmable Logic and Applications (FPL\u201917)","author":"Ma Y.","unstructured":"Y. Ma , Y. Cao , S. Vrudhula , and J. S. Seo . 2017. An automatic RTL compiler for high-throughput FPGA implementation of diverse deep convolutional neural networks . In Proceedings of the 27th International Conference on Field Programmable Logic and Applications (FPL\u201917) . 1--8. Y. Ma, Y. Cao, S. Vrudhula, and J. S. Seo. 2017. An automatic RTL compiler for high-throughput FPGA implementation of diverse deep convolutional neural networks. In Proceedings of the 27th International Conference on Field Programmable Logic and Applications (FPL\u201917). 1--8."},{"key":"e_1_2_1_23_1","doi-asserted-by":"publisher","DOI":"10.1145\/2903150.2911715"},{"key":"e_1_2_1_24_1","doi-asserted-by":"publisher","DOI":"10.1038\/nature14236"},{"key":"e_1_2_1_25_1","doi-asserted-by":"crossref","unstructured":"P. G. Mousouliotis and L. P. Petrou. 2018. SqueezeJet: High-level synthesis accelerator design for deep convolutional neural networks. ArXiv e-prints arxiv:cs.CV\/1805.08695.  P. G. Mousouliotis and L. P. Petrou. 2018. SqueezeJet: High-level synthesis accelerator design for deep convolutional neural networks. ArXiv e-prints arxiv:cs.CV\/1805.08695.","DOI":"10.1007\/978-3-319-78890-6_5"},{"key":"e_1_2_1_26_1","unstructured":"Movidius. 2017. Movidius Neural Compute Stick: Accelerate Deep Learning Development at the Edge. Retrieved from https:\/\/developer.movidius.com\/.  Movidius. 2017. Movidius Neural Compute Stick: Accelerate Deep Learning Development at the Edge. Retrieved from https:\/\/developer.movidius.com\/."},{"key":"e_1_2_1_27_1","unstructured":"NVIDIA. 2017. NVIDIA Deep Learning Accelerator (NVDLA). Retrieved from http:\/\/nvdla.org\/.  NVIDIA. 2017. NVIDIA Deep Learning Accelerator (NVDLA). Retrieved from http:\/\/nvdla.org\/."},{"key":"e_1_2_1_28_1","unstructured":"NVIDIA. 2017. NVIDIA Tegra K1. Retrieved from http:\/\/www.nvidia.com\/object\/tegra-k1-processor.html.  NVIDIA. 2017. NVIDIA Tegra K1. Retrieved from http:\/\/www.nvidia.com\/object\/tegra-k1-processor.html."},{"key":"e_1_2_1_29_1","unstructured":"NVIDIA. 2017. NVIDIA Tegra X1. Retrieved from http:\/\/www.nvidia.com\/object\/tegra-x1-processor.html.  NVIDIA. 2017. NVIDIA Tegra X1. Retrieved from http:\/\/www.nvidia.com\/object\/tegra-x1-processor.html."},{"key":"e_1_2_1_30_1","volume-title":"Proceedings of the 27th International Conference on Field Programmable Logic and Applications (FPL\u201917)","author":"Prost-Boucle A.","unstructured":"A. Prost-Boucle , A. Bourge , F. Petrot , H. Alemdar , N. Caldwell , and V. Leroy . 2017. Scalable high-performance architecture for convolutional ternary neural networks on FPGA . In Proceedings of the 27th International Conference on Field Programmable Logic and Applications (FPL\u201917) . 1--7. A. Prost-Boucle, A. Bourge, F. Petrot, H. Alemdar, N. Caldwell, and V. Leroy. 2017. Scalable high-performance architecture for convolutional ternary neural networks on FPGA. In Proceedings of the 27th International Conference on Field Programmable Logic and Applications (FPL\u201917). 1--7."},{"key":"e_1_2_1_31_1","doi-asserted-by":"publisher","DOI":"10.1145\/2847263.2847265"},{"key":"e_1_2_1_32_1","volume-title":"Proceedings of the Design, Automation Test in Europe. 1--6.","author":"Rahimi A.","unstructured":"A. Rahimi , I. Loi , M. R. Kakoee , and L. Benini . 2011. A fully synthesizable single-cycle interconnection network for Shared-L1 processor clusters . In Proceedings of the Design, Automation Test in Europe. 1--6. A. Rahimi, I. Loi, M. R. Kakoee, and L. Benini. 2011. A fully synthesizable single-cycle interconnection network for Shared-L1 processor clusters. In Proceedings of the Design, Automation Test in Europe. 1--6."},{"key":"e_1_2_1_33_1","doi-asserted-by":"publisher","DOI":"10.1145\/2597917.2597922"},{"key":"e_1_2_1_34_1","unstructured":"K. Simonyan and A. Zisserman. 2014. Very deep convolutional networks for large-scale image recognition. arXiv preprint arXiv:1409.1556.  K. Simonyan and A. Zisserman. 2014. Very deep convolutional networks for large-scale image recognition. arXiv preprint arXiv:1409.1556."},{"key":"e_1_2_1_35_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2014.220"},{"key":"e_1_2_1_36_1","doi-asserted-by":"publisher","DOI":"10.1145\/3020078.3021744"},{"key":"e_1_2_1_37_1","volume-title":"Proceedings of the 27th International Conference on Field Programmable Logic and Applications (FPL\u201917)","author":"Venieris S. I.","unstructured":"S. I. Venieris and C. S. Bouganis . 2017. Latency-driven design for FPGA-based convolutional neural networks . In Proceedings of the 27th International Conference on Field Programmable Logic and Applications (FPL\u201917) . 1--8. S. I. Venieris and C. S. Bouganis. 2017. Latency-driven design for FPGA-based convolutional neural networks. In Proceedings of the 27th International Conference on Field Programmable Logic and Applications (FPL\u201917). 1--8."},{"key":"e_1_2_1_38_1","unstructured":"W. Ren Y. Shengen S. Yi D. Qingqing and S. Gang. 2015. Deep image: Scaling up image recognition. arXiv:1501.02876.  W. Ren Y. Shengen S. Yi D. Qingqing and S. Gang. 2015. Deep image: Scaling up image recognition. arXiv:1501.02876."},{"key":"e_1_2_1_39_1","unstructured":"J. Weston. 2016. Dialog-based language learning. ArXiv:1604.06045 arxiv:cs.CL\/1604.06045.  J. Weston. 2016. Dialog-based language learning. ArXiv:1604.06045 arxiv:cs.CL\/1604.06045."},{"key":"e_1_2_1_40_1","unstructured":"J. Weston S. Chopra and A. Bordes. 2014. Memory networks. ArXiv:1410.3916 arxiv:cs.AI\/1410.3916.  J. Weston S. Chopra and A. Bordes. 2014. Memory networks. ArXiv:1410.3916 arxiv:cs.AI\/1410.3916."},{"key":"e_1_2_1_41_1","unstructured":"Xilinx. 2017. Xilinx Zynq-7000 All Programmable SoC. Retrieved from https:\/\/www.xilinx.com\/products\/silicon-devices\/soc\/zynq-7000.html.  Xilinx. 2017. Xilinx Zynq-7000 All Programmable SoC. Retrieved from https:\/\/www.xilinx.com\/products\/silicon-devices\/soc\/zynq-7000.html."},{"key":"e_1_2_1_42_1","unstructured":"Xilinx. 2017. Zynq UltraScale+ All Programmable Heterogeneous MPSoC. Retrieved from https:\/\/www.xilinx.com\/products\/silicon-devices\/soc\/zynq-ultrascale-mpsoc.html.  Xilinx. 2017. Zynq UltraScale+ All Programmable Heterogeneous MPSoC. Retrieved from https:\/\/www.xilinx.com\/products\/silicon-devices\/soc\/zynq-ultrascale-mpsoc.html."},{"key":"e_1_2_1_43_1","doi-asserted-by":"publisher","DOI":"10.1016\/S0262-4079(16)30458-4"},{"key":"e_1_2_1_44_1","doi-asserted-by":"publisher","DOI":"10.1145\/2966986.2967011"},{"key":"e_1_2_1_45_1","doi-asserted-by":"publisher","DOI":"10.1145\/2684746.2689060"}],"container-title":["ACM Transactions on Reconfigurable Technology and Systems"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3284357","content-type":"unspecified","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3284357","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,6,18]],"date-time":"2025-06-18T00:43:49Z","timestamp":1750207429000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3284357"}},"subtitle":["Exploiting CPU-FPGA Synergies for Efficient and Flexible CNN Inference Acceleration on Zynq SoCs"],"short-title":[],"issued":{"date-parts":[[2018,9,30]]},"references-count":45,"journal-issue":{"issue":"3","published-print":{"date-parts":[[2018,9,30]]}},"alternative-id":["10.1145\/3284357"],"URL":"https:\/\/doi.org\/10.1145\/3284357","relation":{},"ISSN":["1936-7406","1936-7414"],"issn-type":[{"value":"1936-7406","type":"print"},{"value":"1936-7414","type":"electronic"}],"subject":[],"published":{"date-parts":[[2018,9,30]]},"assertion":[{"value":"2017-12-01","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2018-08-01","order":1,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2018-12-12","order":2,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}