{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,10]],"date-time":"2026-03-10T15:15:38Z","timestamp":1773155738283,"version":"3.50.1"},"reference-count":54,"publisher":"Association for Computing Machinery (ACM)","issue":"4","license":[{"start":{"date-parts":[[2021,7,19]],"date-time":"2021-07-19T00:00:00Z","timestamp":1626652800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/www.acm.org\/publications\/policies\/copyright_policy#Background"}],"funder":[{"name":"Okawa Research grant"},{"DOI":"10.13039\/100000001","name":"NSF","doi-asserted-by":"publisher","award":["1553333"],"award-info":[{"award-number":["1553333"]}],"id":[{"id":"10.13039\/100000001","id-type":"DOI","asserted-by":"publisher"}]},{"name":"Sloan Fellowship"},{"name":"Swiss National Foundation","award":["P2EZP2 181817"],"award-info":[{"award-number":["P2EZP2 181817"]}]},{"name":"PECASE"}],"content-domain":{"domain":["dl.acm.org"],"crossmark-restriction":true},"short-container-title":["ACM Trans. Graph."],"published-print":{"date-parts":[[2021,8,31]]},"abstract":"<jats:p>Neural representations have emerged as a new paradigm for applications in rendering, imaging, geometric modeling, and simulation. Compared to traditional representations such as meshes, point clouds, or volumes they can be flexibly incorporated into differentiable learning-based pipelines. While recent improvements to neural representations now make it possible to represent signals with fine details at moderate resolutions (e.g., for images and 3D shapes), adequately representing large-scale or complex scenes has proven a challenge. Current neural representations fail to accurately represent images at resolutions greater than a megapixel or 3D scenes with more than a few hundred thousand polygons. Here, we introduce a new hybrid implicit-explicit network architecture and training strategy that adaptively allocates resources during training and inference based on the local complexity of a signal of interest. Our approach uses a multiscale block-coordinate decomposition, similar to a quadtree or octree, that is optimized during training. The network architecture operates in two stages: using the bulk of the network parameters, a coordinate encoder generates a feature grid in a single forward pass. Then, hundreds or thousands of samples within each block can be efficiently evaluated using a lightweight feature decoder. With this hybrid implicit-explicit network architecture, we demonstrate the first experiments that fit gigapixel images to nearly 40 dB peak signal-to-noise ratio. Notably this represents an increase in scale of over 1000X compared to the resolution of previously demonstrated image-fitting experiments. Moreover, our approach is able to represent 3D shapes significantly faster and better than previous techniques; it reduces training times from days to hours or minutes and memory requirements by over an order of magnitude.<\/jats:p>","DOI":"10.1145\/3450626.3459785","type":"journal-article","created":{"date-parts":[[2021,7,20]],"date-time":"2021-07-20T00:04:27Z","timestamp":1626739467000},"page":"1-13","update-policy":"https:\/\/doi.org\/10.1145\/crossmark-policy","source":"Crossref","is-referenced-by-count":160,"title":["Acorn"],"prefix":"10.1145","volume":"40","author":[{"given":"Julien N. P.","family":"Martel","sequence":"first","affiliation":[{"name":"Stanford University"}]},{"given":"David B.","family":"Lindell","sequence":"additional","affiliation":[{"name":"Stanford University"}]},{"given":"Connor Z.","family":"Lin","sequence":"additional","affiliation":[{"name":"Stanford University"}]},{"given":"Eric R.","family":"Chan","sequence":"additional","affiliation":[{"name":"Stanford University"}]},{"given":"Marco","family":"Monteiro","sequence":"additional","affiliation":[{"name":"Stanford University"}]},{"given":"Gordon","family":"Wetzstein","sequence":"additional","affiliation":[{"name":"Stanford University"}]}],"member":"320","published-online":{"date-parts":[[2021,7,19]]},"reference":[{"key":"e_1_2_2_1_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-58452-8_26"},{"key":"e_1_2_2_2_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.00264"},{"key":"e_1_2_2_3_1","doi-asserted-by":"publisher","DOI":"10.1016\/0021-9991(84)90073-1"},{"key":"e_1_2_2_4_1","doi-asserted-by":"publisher","DOI":"10.1145\/3386569.3392485"},{"key":"e_1_2_2_5_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-58526-6_36"},{"key":"e_1_2_2_6_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.00574"},{"key":"e_1_2_2_7_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.00852"},{"key":"e_1_2_2_8_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00609"},{"key":"e_1_2_2_9_1","volume-title":"Overfit neural networks as a compact shape representation. arXiv preprint arXiv:2009.09808","author":"Davies Thomas","year":"2020","unstructured":"Thomas Davies , Derek Nowrouzezahrai , and Alec Jacobson . 2020. Overfit neural networks as a compact shape representation. arXiv preprint arXiv:2009.09808 ( 2020 ). Thomas Davies, Derek Nowrouzezahrai, and Alec Jacobson. 2020. Overfit neural networks as a compact shape representation. arXiv preprint arXiv:2009.09808 (2020)."},{"key":"e_1_2_2_10_1","doi-asserted-by":"crossref","unstructured":"S. M. Ali Eslami Danilo Jimenez Rezende Frederic Besse Fabio Viola Ari S. Morcos Marta Garnelo Avraham Ruderman Andrei A. Rusu Ivo Danihelka Karol Gregor David P. Reichert Lars Buesing Theophane Weber Oriol Vinyals Dan Rosenbaum Neil Rabinowitz Helen King Chloe Hillier Matt Botvinick Daan Wierstra Koray Kavukcuoglu and Demis Hassabis. 2018. Neural scene representation and rendering. Science 360 6394 (2018) 1204--1210.  S. M. Ali Eslami Danilo Jimenez Rezende Frederic Besse Fabio Viola Ari S. Morcos Marta Garnelo Avraham Ruderman Andrei A. Rusu Ivo Danihelka Karol Gregor David P. Reichert Lars Buesing Theophane Weber Oriol Vinyals Dan Rosenbaum Neil Rabinowitz Helen King Chloe Hillier Matt Botvinick Daan Wierstra Koray Kavukcuoglu and Demis Hassabis. 2018. Neural scene representation and rendering. Science 360 6394 (2018) 1204--1210.","DOI":"10.1126\/science.aar6170"},{"key":"e_1_2_2_11_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00247"},{"key":"e_1_2_2_12_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.00491"},{"key":"e_1_2_2_13_1","volume-title":"Proc. ICML.","author":"Gropp Amos","year":"2020","unstructured":"Amos Gropp , Lior Yariv , Niv Haim , Matan Atzmon , and Yaron Lipman . 2020 . Implicit geometric regularization for learning shapes . In Proc. ICML. Amos Gropp, Lior Yariv, Niv Haim, Matan Atzmon, and Yaron Lipman. 2020. Implicit geometric regularization for learning shapes. In Proc. ICML."},{"key":"e_1_2_2_14_1","unstructured":"LLC Gurobi Optimization. 2021. Gurobi Optimizer Reference Manual. http:\/\/www.gurobi.com  LLC Gurobi Optimization. 2021. Gurobi Optimizer Reference Manual. http:\/\/www.gurobi.com"},{"key":"e_1_2_2_15_1","volume-title":"Dally","author":"Han Song","year":"2016","unstructured":"Song Han , Huizi Mao , and William J . Dally . 2016 . Deep compression: Compressing deep neural networks with pruning, trained quantization and Huffman coding. In Proc. ICLR. Song Han, Huizi Mao, and William J. Dally. 2016. Deep compression: Compressing deep neural networks with pruning, trained quantization and Huffman coding. In Proc. ICLR."},{"key":"e_1_2_2_16_1","article-title":"Deep blending for free-viewpoint image-based rendering","volume":"37","author":"Hedman Peter","year":"2018","unstructured":"Peter Hedman , Julien Philip , True Price , Jan-Michael Frahm , George Drettakis , and Gabriel Brostow . 2018 . Deep blending for free-viewpoint image-based rendering . ACM Trans. Graph. (SIGGRAPH Asia) 37 , 6 (2018). Peter Hedman, Julien Philip, True Price, Jan-Michael Frahm, George Drettakis, and Gabriel Brostow. 2018. Deep blending for free-viewpoint image-based rendering. ACM Trans. Graph. (SIGGRAPH Asia) 37, 6 (2018).","journal-title":"ACM Trans. Graph. (SIGGRAPH Asia)"},{"key":"e_1_2_2_17_1","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.01008"},{"key":"e_1_2_2_18_1","volume-title":"Russell","author":"Huang Weizhang","year":"2010","unstructured":"Weizhang Huang and Robert D . Russell . 2010 . Adaptive Moving Mesh Methods. Springer New York . Weizhang Huang and Robert D. Russell. 2010. Adaptive Moving Mesh Methods. Springer New York."},{"key":"e_1_2_2_19_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.00604"},{"key":"e_1_2_2_20_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.00133"},{"key":"e_1_2_2_21_1","doi-asserted-by":"publisher","DOI":"10.1145\/2487228.2487237"},{"key":"e_1_2_2_22_1","doi-asserted-by":"crossref","unstructured":"Petr Kellnhofer Lars Jebe Andrew Jones Ryan Spicer Kari Pulli and Gordon Wetzstein. 2021. Neural Lumigraph Rendering. In CVPR.  Petr Kellnhofer Lars Jebe Andrew Jones Ryan Spicer Kari Pulli and Gordon Wetzstein. 2021. Neural Lumigraph Rendering. In CVPR.","DOI":"10.1109\/CVPR46437.2021.00427"},{"key":"e_1_2_2_23_1","doi-asserted-by":"publisher","DOI":"10.1111\/cgf.13619"},{"key":"e_1_2_2_24_1","volume-title":"Kingma and Jimmy Ba","author":"Diederik","year":"2014","unstructured":"Diederik P. Kingma and Jimmy Ba . 2014 . Adam : A Method for Stochastic Optimization. In Proc. ICLR. Diederik P. Kingma and Jimmy Ba. 2014. Adam: A Method for Stochastic Optimization. In Proc. ICLR."},{"key":"e_1_2_2_25_1","doi-asserted-by":"publisher","DOI":"10.1109\/3DV50981.2020.00052"},{"key":"e_1_2_2_26_1","unstructured":"Stanford Computer Graphics Laboratory. 2014. Stanford 3D Scanning Repository. http:\/\/graphics.stanford.edu\/data\/3Dscanrep\/  Stanford Computer Graphics Laboratory. 2014. Stanford 3D Scanning Repository. http:\/\/graphics.stanford.edu\/data\/3Dscanrep\/"},{"key":"e_1_2_2_27_1","volume-title":"Proc. NeurIPS.","author":"Li Zongyi","year":"2020","unstructured":"Zongyi Li , Nikola Kovachki , Kamyar Azizzadenesheli , Burigede Liu , Kaushik Bhattacharya , Andrew Stuart , and Anima Anandkumar . 2020 . Fourier neural operator for parametric partial differential equations . In Proc. NeurIPS. Zongyi Li, Nikola Kovachki, Kamyar Azizzadenesheli, Burigede Liu, Kaushik Bhattacharya, Andrew Stuart, and Anima Anandkumar. 2020. Fourier neural operator for parametric partial differential equations. In Proc. NeurIPS."},{"key":"e_1_2_2_28_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 a. Neural sparse voxel fields. In NeurIPS. Lingjie Liu, Jiatao Gu, Kyaw Zaw Lin, Tat-Seng Chua, and Christian Theobalt. 2020a. Neural sparse voxel fields. In NeurIPS."},{"key":"e_1_2_2_29_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.00209"},{"key":"e_1_2_2_30_1","doi-asserted-by":"publisher","DOI":"10.1145\/3306346.3323020"},{"key":"e_1_2_2_31_1","doi-asserted-by":"publisher","DOI":"10.1145\/37402.37422"},{"key":"e_1_2_2_32_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00459"},{"key":"e_1_2_2_33_1","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00484"},{"key":"e_1_2_2_34_1","doi-asserted-by":"publisher","DOI":"10.1145\/3306346.3322980"},{"key":"e_1_2_2_35_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-58452-8_24"},{"key":"e_1_2_2_36_1","volume-title":"Proc. NeurIPS.","author":"Nguyen-Phuoc Thu","year":"2020","unstructured":"Thu Nguyen-Phuoc , Christian Richardt , Long Mai , Yong-Liang Yang , and Niloy Mitra . 2020 . BlockGAN: Learning 3D object-aware scene representations from unlabelled images . In Proc. NeurIPS. Thu Nguyen-Phuoc, Christian Richardt, Long Mai, Yong-Liang Yang, and Niloy Mitra. 2020. BlockGAN: Learning 3D object-aware scene representations from unlabelled images. In Proc. NeurIPS."},{"key":"e_1_2_2_37_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.00356"},{"key":"e_1_2_2_38_1","doi-asserted-by":"crossref","unstructured":"Michael Niemeyer Lars Mescheder Michael Oechsle and Andreas Geiger. 2020b. Differentiable volumetric rendering: Learning implicit 3d representations without 3d supervision. In CVPR.  Michael Niemeyer Lars Mescheder Michael Oechsle and Andreas Geiger. 2020b. Differentiable volumetric rendering: Learning implicit 3d representations without 3d supervision. In CVPR.","DOI":"10.1109\/CVPR42600.2020.00356"},{"key":"e_1_2_2_39_1","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00463"},{"key":"e_1_2_2_40_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00025"},{"key":"e_1_2_2_41_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-58580-8_31"},{"key":"e_1_2_2_42_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-58529-7_37"},{"key":"e_1_2_2_43_1","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00239"},{"key":"e_1_2_2_44_1","volume-title":"Proc. NeurIPS.","author":"Schwarz Katja","year":"2020","unstructured":"Katja Schwarz , Yiyi Liao , Michael Niemeyer , and Andreas Geiger . 2020 . GRAF: Generative radiance fields for 3D-aware image synthesis . In Proc. NeurIPS. Katja Schwarz, Yiyi Liao, Michael Niemeyer, and Andreas Geiger. 2020. GRAF: Generative radiance fields for 3D-aware image synthesis. In Proc. NeurIPS."},{"key":"e_1_2_2_45_1","volume-title":"Proc. NeurIPS.","author":"Sitzmann Vincent","year":"2020","unstructured":"Vincent Sitzmann , Julien N. P. Martel , Alexander W. Bergman , David B. Lindell , and Gordon Wetzstein . 2020 . Implicit neural representations with periodic activation functions . In Proc. NeurIPS. Vincent Sitzmann, Julien N. P. Martel, Alexander W. Bergman, David B. Lindell, and Gordon Wetzstein. 2020. Implicit neural representations with periodic activation functions. In Proc. NeurIPS."},{"key":"e_1_2_2_46_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00254"},{"key":"e_1_2_2_47_1","volume-title":"Proc. NeurIPS.","author":"Sitzmann Vincent","year":"2019","unstructured":"Vincent Sitzmann , Michael Zollh\u00f6fer , and Gordon Wetzstein . 2019 b. Scene representation networks: Continuous 3D-structure-aware neural scene representations . In Proc. NeurIPS. Vincent Sitzmann, Michael Zollh\u00f6fer, and Gordon Wetzstein. 2019b. Scene representation networks: Continuous 3D-structure-aware neural scene representations. In Proc. NeurIPS."},{"key":"e_1_2_2_48_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.01120"},{"key":"e_1_2_2_49_1","volume-title":"Proc. NeurIPS","author":"Tancik Matthew","year":"2020","unstructured":"Matthew Tancik , Pratul P. Srinivasan , Ben Mildenhall , Sara Fridovich-Keil , Nithin Raghavan , Utkarsh Singhal , Ravi Ramamoorthi , Jonathan T. Barron , and Ren Ng . 2020 . Fourier features let networks learn high frequency functions in low dimensional domains . Proc. NeurIPS (2020). Matthew Tancik, Pratul P. Srinivasan, Ben Mildenhall, Sara Fridovich-Keil, Nithin Raghavan, Utkarsh Singhal, Ravi Ramamoorthi, Jonathan T. Barron, and Ren Ng. 2020. Fourier features let networks learn high frequency functions in low dimensional domains. Proc. NeurIPS (2020)."},{"key":"e_1_2_2_50_1","doi-asserted-by":"publisher","DOI":"10.1111\/cgf.14022"},{"key":"e_1_2_2_51_1","doi-asserted-by":"publisher","DOI":"10.1145\/3306346.3323035"},{"key":"e_1_2_2_52_1","volume-title":"Proc. NeurIPS.","author":"Yariv Lior","year":"2020","unstructured":"Lior Yariv , Yoni Kasten , Dror Moran , Meirav Galun , Matan Atzmon , Ronen Basri , and Yaron Lipman . 2020 . Multiview neural surface reconstruction by disentangling geometry and appearance . In Proc. NeurIPS. Lior Yariv, Yoni Kasten, Dror Moran, Meirav Galun, Matan Atzmon, Ronen Basri, and Yaron Lipman. 2020. Multiview neural surface reconstruction by disentangling geometry and appearance. In Proc. NeurIPS."},{"key":"e_1_2_2_53_1","doi-asserted-by":"publisher","DOI":"10.1145\/3446328"},{"key":"e_1_2_2_54_1","doi-asserted-by":"publisher","DOI":"10.1145\/3197517.3201323"}],"container-title":["ACM Transactions on Graphics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3450626.3459785","content-type":"unspecified","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3450626.3459785","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3450626.3459785","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,6,17]],"date-time":"2025-06-17T20:17:16Z","timestamp":1750191436000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3450626.3459785"}},"subtitle":["adaptive coordinate networks for neural scene representation"],"short-title":[],"issued":{"date-parts":[[2021,7,19]]},"references-count":54,"aliases":["10.1145\/3476576.3476607"],"journal-issue":{"issue":"4","published-print":{"date-parts":[[2021,8,31]]}},"alternative-id":["10.1145\/3450626.3459785"],"URL":"https:\/\/doi.org\/10.1145\/3450626.3459785","relation":{},"ISSN":["0730-0301","1557-7368"],"issn-type":[{"value":"0730-0301","type":"print"},{"value":"1557-7368","type":"electronic"}],"subject":[],"published":{"date-parts":[[2021,7,19]]},"assertion":[{"value":"2021-07-19","order":2,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}