{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,11]],"date-time":"2026-03-11T01:42:47Z","timestamp":1773193367925,"version":"3.50.1"},"reference-count":59,"publisher":"Association for Computing Machinery (ACM)","issue":"6","license":[{"start":{"date-parts":[[2022,11,30]],"date-time":"2022-11-30T00:00:00Z","timestamp":1669766400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["dl.acm.org"],"crossmark-restriction":true},"short-container-title":["ACM Trans. Graph."],"published-print":{"date-parts":[[2022,12]]},"abstract":"<jats:p>\n            The practical deployment of Neural Radiance Fields (NeRF) in rendering applications faces several challenges, with the most critical one being low rendering speed on even high-end graphic processing units (GPUs). In this paper, we present ICARUS, a specialized accelerator architecture tailored for NeRF rendering. Unlike GPUs using general purpose computing and memory architectures for NeRF, ICARUS executes the complete NeRF pipeline using dedicated plenoptic cores (PLCore) consisting of a positional encoding unit (PEU), a multi-layer perceptron (MLP) engine, and a volume rendering unit (VRU). A PLCore takes in positions &amp; directions and renders the corresponding pixel colors without any intermediate data going off-chip for temporary storage and exchange, which can be time and power consuming. To implement the most expensive component of NeRF, i.e., the MLP, we transform the fully connected operations to approximated reconfigurable multiple constant multiplications (MCMs), where common subexpressions are shared across different multiplications to improve the computation efficiency. We build a prototype ICARUS using Synopsys HAPS-80 S104, a field programmable gate array (FPGA)-based prototyping system for large-scale integrated circuits and systems design. We evaluate the power-performancearea (PPA) of a PLCore using 40nm LP CMOS technology. Working at 400 MHz, a single PLCore occupies 16.5\n            <jats:italic>mm<\/jats:italic>\n            <jats:sup>2<\/jats:sup>\n            and consumes 282.8 mW, translating to 0.105 uJ\/sample. The results are compared with those of GPU and tensor processing unit (TPU) implementations.\n          <\/jats:p>","DOI":"10.1145\/3550454.3555505","type":"journal-article","created":{"date-parts":[[2022,11,30]],"date-time":"2022-11-30T21:19:07Z","timestamp":1669843147000},"page":"1-14","update-policy":"https:\/\/doi.org\/10.1145\/crossmark-policy","source":"Crossref","is-referenced-by-count":47,"title":["ICARUS"],"prefix":"10.1145","volume":"41","author":[{"given":"Chaolin","family":"Rao","sequence":"first","affiliation":[{"name":"ShanghaiTech University, China and GGU Technology Co., Ltd., China"}]},{"given":"Huangjie","family":"Yu","sequence":"additional","affiliation":[{"name":"ShanghaiTech University, China"}]},{"given":"Haochuan","family":"Wan","sequence":"additional","affiliation":[{"name":"ShanghaiTech University, China"}]},{"given":"Jindong","family":"Zhou","sequence":"additional","affiliation":[{"name":"ShanghaiTech University, China"}]},{"given":"Yueyang","family":"Zheng","sequence":"additional","affiliation":[{"name":"ShanghaiTech University, China"}]},{"given":"Minye","family":"Wu","sequence":"additional","affiliation":[{"name":"KU Leuven, Belgium"}]},{"given":"Yu","family":"Ma","sequence":"additional","affiliation":[{"name":"ShanghaiTech University, China"}]},{"given":"Anpei","family":"Chen","sequence":"additional","affiliation":[{"name":"ShanghaiTech University, China"}]},{"given":"Binzhe","family":"Yuan","sequence":"additional","affiliation":[{"name":"ShanghaiTech University, China"}]},{"given":"Pingqiang","family":"Zhou","sequence":"additional","affiliation":[{"name":"ShanghaiTech University, China"}]},{"given":"Xin","family":"Lou","sequence":"additional","affiliation":[{"name":"ShanghaiTech University, China and GGU Technology Co., Ltd., China"}]},{"given":"Jingyi","family":"Yu","sequence":"additional","affiliation":[{"name":"ShanghaiTech University, China"}]}],"member":"320","published-online":{"date-parts":[[2022,11,30]]},"reference":[{"key":"e_1_2_1_1_1","volume-title":"Real-time rendering","author":"Akenine-M\u00f6ller Tomas","unstructured":"Tomas Akenine-M\u00f6ller, Eric Haines, and Naty Hoffman. 2019. Real-time rendering. Crc Press."},{"key":"e_1_2_1_2_1","volume-title":"Glsl geometry shaders","author":"Bailey Mike","year":"2007","unstructured":"Mike Bailey. 2007. Glsl geometry shaders. Oregon State University (2007)."},{"key":"e_1_2_1_3_1","doi-asserted-by":"publisher","DOI":"10.1145\/383259.383309"},{"key":"e_1_2_1_4_1","volume-title":"A subdivision algorithm for computer display of curved surfaces","author":"Catmull Edwin Earl","unstructured":"Edwin Earl Catmull. 1974. A subdivision algorithm for computer display of curved surfaces. The University of Utah."},{"key":"e_1_2_1_5_1","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1145\/3203192","article-title":"Deep surface light fields","volume":"1","author":"Chen Anpei","year":"2018","unstructured":"Anpei Chen, Minye Wu, Yingliang Zhang, Nianyi Li, Jie Lu, Shenghua Gao, and Jingyi Yu. 2018. Deep surface light fields. Proceedings of the ACM on Computer Graphics and Interactive Techniques 1, 1 (2018), 1--17.","journal-title":"Proceedings of the ACM on Computer Graphics and Interactive Techniques"},{"key":"e_1_2_1_6_1","doi-asserted-by":"publisher","DOI":"10.1145\/2654822.2541967"},{"key":"e_1_2_1_7_1","doi-asserted-by":"publisher","DOI":"10.1109\/MICRO.2014.58"},{"key":"e_1_2_1_8_1","doi-asserted-by":"publisher","DOI":"10.1109\/JSSC.2016.2616357"},{"key":"e_1_2_1_9_1","doi-asserted-by":"publisher","DOI":"10.1109\/JETCAS.2019.2910232"},{"key":"e_1_2_1_10_1","volume-title":"Srinivasan","author":"Deng Boyang","year":"2020","unstructured":"Boyang Deng, Jonathan T. Barron, and Pratul P. Srinivasan. 2020. JaxNeRF: an efficient JAX implementation of NeRF. https:\/\/github.com\/google-research\/google-research\/tree\/master\/jaxnerf."},{"key":"e_1_2_1_11_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01254"},{"key":"e_1_2_1_12_1","doi-asserted-by":"publisher","DOI":"10.1145\/2749469.2750389"},{"key":"e_1_2_1_13_1","volume-title":"SIGNET: Efficient Neural Representations for Light Fields. In International Conference on Computer Vision (ICCV","author":"Brandon","year":"2021","unstructured":"Brandon Y. Feng and Amitabh Varshney. 2021. SIGNET: Efficient Neural Representations for Light Fields. In International Conference on Computer Vision (ICCV 2021)."},{"key":"e_1_2_1_14_1","volume-title":"Fastnerf: High-fidelity neural rendering at 200fps. arXiv preprint arXiv:2103.10380","author":"Garbin Stephan J","year":"2021","unstructured":"Stephan J Garbin, Marek Kowalski, Matthew Johnson, Jamie Shotton, and Julien Valentin. 2021. Fastnerf: High-fidelity neural rendering at 200fps. arXiv preprint arXiv:2103.10380 (2021)."},{"key":"e_1_2_1_15_1","volume-title":"Tessellating and Rendering Bezier\/B-Spline\/NURBS Curves and Surfaces using Geometry Shader in GPU","author":"Han Dongsoo","unstructured":"Dongsoo Han. 2007. Tessellating and Rendering Bezier\/B-Spline\/NURBS Curves and Surfaces using Geometry Shader in GPU. University of Pennsylvania (2007)."},{"key":"e_1_2_1_16_1","volume-title":"AMD to Introduce New Next-Gen RDNA GPUs","author":"Tom's","year":"2020","unstructured":"Tom's hardware. 2020. AMD to Introduce New Next-Gen RDNA GPUs in 2020, Not a Typical 'Refresh' of Navi. http:\/\/https:\/\/www.tomshardware.com\/news\/amds-navi-to-be-refreshed-with-next-gen-rdna-architecture-in-2020"},{"key":"e_1_2_1_17_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00286"},{"key":"e_1_2_1_18_1","doi-asserted-by":"publisher","DOI":"10.1109\/JSSC.2017.2764045"},{"key":"e_1_2_1_19_1","doi-asserted-by":"publisher","DOI":"10.1145\/3079856.3080246"},{"key":"e_1_2_1_20_1","doi-asserted-by":"crossref","unstructured":"J. T. Kajiya and H. B. Von. 1984. Ray tracing volume densities\" computer graphics 18. (1984).","DOI":"10.1145\/964965.808594"},{"key":"e_1_2_1_21_1","doi-asserted-by":"publisher","DOI":"10.1145\/1294685.1294699"},{"key":"e_1_2_1_22_1","doi-asserted-by":"publisher","DOI":"10.1109\/JSSC.2011.2171417"},{"key":"e_1_2_1_23_1","doi-asserted-by":"publisher","DOI":"10.1109\/TCSI.2012.2209302"},{"key":"e_1_2_1_24_1","doi-asserted-by":"publisher","DOI":"10.1145\/2694344.2694358"},{"key":"e_1_2_1_25_1","volume-title":"Tat-Seng Chua, and Christian Theobalt.","author":"Liu Lingjie","year":"2020","unstructured":"Lingjie Liu, Jiatao Gu, Kyaw Zaw Lin, Tat-Seng Chua, and Christian Theobalt. 2020. Neural sparse voxel fields. arXiv preprint arXiv:2007.11571 (2020)."},{"key":"e_1_2_1_26_1","volume-title":"Convolutional Neural Opacity Radiance Fields. arXiv preprint arXiv:2104.01772","author":"Luo Haimin","year":"2021","unstructured":"Haimin Luo, Anpei Chen, Qixuan Zhang, Bai Pang, Minye Wu, Lan Xu, and Jingyi Yu. 2021. Convolutional Neural Opacity Radiance Fields. arXiv preprint arXiv:2104.01772 (2021)."},{"key":"e_1_2_1_27_1","doi-asserted-by":"publisher","DOI":"10.1109\/2945.468400"},{"key":"e_1_2_1_28_1","volume-title":"A Survey on Bounding","author":"Meister Daniel","unstructured":"Daniel Meister, Shinji Ogaki, Carsten Benthin, Michael J Doyle, Michael Guthe, and Ji\u0159\u00ed Bittner. 2021. A Survey on Bounding Volume Hierarchies for Ray Tracing. In Computer Graphics Forum, Vol. 40. Wiley Online Library, 683--712."},{"key":"e_1_2_1_29_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-58452-8_24"},{"key":"e_1_2_1_30_1","doi-asserted-by":"publisher","DOI":"10.1109\/JSSC.2016.2636225"},{"key":"e_1_2_1_31_1","volume-title":"Instant Neural Graphics Primitives with a Multiresolution Hash Encoding. arXiv preprint arXiv:2201.05989","author":"M\u00fcller Thomas","year":"2022","unstructured":"Thomas M\u00fcller, Alex Evans, Christoph Schied, and Alexander Keller. 2022. Instant Neural Graphics Primitives with a Multiresolution Hash Encoding. arXiv preprint arXiv:2201.05989 (2022)."},{"key":"e_1_2_1_32_1","doi-asserted-by":"publisher","DOI":"10.1145\/3528223.3530127"},{"key":"e_1_2_1_33_1","doi-asserted-by":"publisher","DOI":"10.1109\/TVCG.2014.2371855"},{"key":"e_1_2_1_34_1","doi-asserted-by":"publisher","DOI":"10.1145\/2629634"},{"key":"e_1_2_1_35_1","doi-asserted-by":"publisher","DOI":"10.1145\/2070781.2024194"},{"key":"e_1_2_1_36_1","doi-asserted-by":"publisher","DOI":"10.1145\/800193.569954"},{"key":"e_1_2_1_37_1","unstructured":"Nvidia. 2020. NVIDIA Ampere Architecture. https:\/\/www.nvidia.com\/en-us\/data-center\/ampere-architecture\/"},{"key":"e_1_2_1_38_1","doi-asserted-by":"publisher","DOI":"10.1109\/JSSC.2003.821785"},{"key":"e_1_2_1_39_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00025"},{"key":"e_1_2_1_40_1","doi-asserted-by":"publisher","DOI":"10.1145\/74333.74359"},{"key":"e_1_2_1_41_1","doi-asserted-by":"publisher","DOI":"10.1145\/311535.311563"},{"key":"e_1_2_1_42_1","volume-title":"International Conference on Machine Learning. PMLR, 5301--5310","author":"Rahaman Nasim","year":"2019","unstructured":"Nasim Rahaman, Aristide Baratin, Devansh Arpit, Felix Draxler, Min Lin, Fred Hamprecht, Yoshua Bengio, and Aaron Courville. 2019. On the spectral bias of neural networks. In International Conference on Machine Learning. PMLR, 5301--5310."},{"key":"e_1_2_1_43_1","volume-title":"KiloNeRF: Speeding up Neural Radiance Fields with Thousands of Tiny MLPs. arXiv preprint arXiv:2103.13744","author":"Reiser Christian","year":"2021","unstructured":"Christian Reiser, Songyou Peng, Yiyi Liao, and Andreas Geiger. 2021. KiloNeRF: Speeding up Neural Radiance Fields with Thousands of Tiny MLPs. arXiv preprint arXiv:2103.13744 (2021)."},{"key":"e_1_2_1_44_1","doi-asserted-by":"publisher","DOI":"10.1134\/S0361768820030068"},{"key":"e_1_2_1_45_1","volume-title":"Proceedings of the ACM SIGGRAPH\/EUROGRAPHICS Conference on Graphics Hardware","author":"Schmittler J\u00f6rg","year":"2002","unstructured":"J\u00f6rg Schmittler, Ingo Wald, and Philipp Slusallek. 2002. SaarCOR: A Hardware Architecture for Ray Tracing. In Proceedings of the ACM SIGGRAPH\/EUROGRAPHICS Conference on Graphics Hardware (Saarbrucken, Germany) (HWWS '02). Eurographics Association, Goslar, DEU, 27--36."},{"key":"e_1_2_1_46_1","doi-asserted-by":"publisher","DOI":"10.1109\/TCAD.2009.2028981"},{"key":"e_1_2_1_47_1","volume-title":"Direct Voxel Grid Optimization: Super-fast Convergence for Radiance Fields Reconstruction. arXiv preprint arXiv:2111.11215","author":"Sun Cheng","year":"2021","unstructured":"Cheng Sun, Min Sun, and Hwann-Tzong Chen. 2021. Direct Voxel Grid Optimization: Super-fast Convergence for Radiance Fields Reconstruction. arXiv preprint arXiv:2111.11215 (2021)."},{"key":"e_1_2_1_48_1","volume-title":"Fourier features let networks learn high frequency functions in low dimensional domains. Advances in Neural Information Processing Systems 33","author":"Tancik Matthew","year":"2020","unstructured":"Matthew Tancik, Pratul Srinivasan, Ben Mildenhall, Sara Fridovich-Keil, Nithin Raghavan, Utkarsh Singhal, Ravi Ramamoorthi, Jonathan Barron, and Ren Ng. 2020. Fourier features let networks learn high frequency functions in low dimensional domains. Advances in Neural Information Processing Systems 33 (2020)."},{"key":"e_1_2_1_49_1","doi-asserted-by":"publisher","DOI":"10.1145\/166117.166152"},{"key":"e_1_2_1_50_1","doi-asserted-by":"publisher","DOI":"10.1145\/344779.344925"},{"key":"e_1_2_1_51_1","volume-title":"A programmable hardware architecture for realtime ray tracing of coherent dynamic scenes. Diss. Ph. D. Thesis","author":"Woop Sven","year":"2007","unstructured":"Sven Woop. 2007. A programmable hardware architecture for realtime ray tracing of coherent dynamic scenes. Diss. Ph. D. Thesis, Sarrland University (2007)."},{"key":"e_1_2_1_52_1","doi-asserted-by":"publisher","DOI":"10.1145\/1073204.1073211"},{"key":"e_1_2_1_53_1","doi-asserted-by":"publisher","DOI":"10.1109\/JSSC.2017.2778281"},{"key":"e_1_2_1_54_1","doi-asserted-by":"publisher","DOI":"10.1109\/VLSIC.2018.8502388"},{"key":"e_1_2_1_55_1","doi-asserted-by":"publisher","DOI":"10.1109\/VLSIC.2018.8502309"},{"key":"e_1_2_1_56_1","volume-title":"Plenoxels: Radiance Fields without Neural Networks. arXiv preprint arXiv:2112.05131","author":"Yu Alex","year":"2021","unstructured":"Alex Yu, Sara Fridovich-Keil, Matthew Tancik, Qinhong Chen, Benjamin Recht, and Angjoo Kanazawa. 2021a. Plenoxels: Radiance Fields without Neural Networks. arXiv preprint arXiv:2112.05131 (2021)."},{"key":"e_1_2_1_57_1","volume-title":"Plenoctrees for real-time rendering of neural radiance fields. arXiv preprint arXiv:2103.14024","author":"Yu Alex","year":"2021","unstructured":"Alex Yu, Ruilong Li, Matthew Tancik, Hao Li, Ren Ng, and Angjoo Kanazawa. 2021b. Plenoctrees for real-time rendering of neural radiance fields. arXiv preprint arXiv:2103.14024 (2021)."},{"key":"e_1_2_1_58_1","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v36i3.20223"},{"key":"e_1_2_1_59_1","volume-title":"Reconstructing continuous distributions of 3D protein structure from cryo-EM images. arXiv preprint arXiv:1909.05215","author":"Zhong Ellen D","year":"2019","unstructured":"Ellen D Zhong, Tristan Bepler, Joseph H Davis, and Bonnie Berger. 2019. Reconstructing continuous distributions of 3D protein structure from cryo-EM images. arXiv preprint arXiv:1909.05215 (2019)."}],"container-title":["ACM Transactions on Graphics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3550454.3555505","content-type":"unspecified","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3550454.3555505","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,6,17]],"date-time":"2025-06-17T17:51:43Z","timestamp":1750182703000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3550454.3555505"}},"subtitle":["A Specialized Architecture for Neural Radiance Fields Rendering"],"short-title":[],"issued":{"date-parts":[[2022,11,30]]},"references-count":59,"journal-issue":{"issue":"6","published-print":{"date-parts":[[2022,12]]}},"alternative-id":["10.1145\/3550454.3555505"],"URL":"https:\/\/doi.org\/10.1145\/3550454.3555505","relation":{},"ISSN":["0730-0301","1557-7368"],"issn-type":[{"value":"0730-0301","type":"print"},{"value":"1557-7368","type":"electronic"}],"subject":[],"published":{"date-parts":[[2022,11,30]]},"assertion":[{"value":"2022-11-30","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}