{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,1,27]],"date-time":"2026-01-27T02:25:12Z","timestamp":1769480712336,"version":"3.49.0"},"reference-count":83,"publisher":"Association for Computing Machinery (ACM)","issue":"3","license":[{"start":{"date-parts":[[2024,4,9]],"date-time":"2024-04-09T00:00:00Z","timestamp":1712620800000},"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":[[2024,6,30]]},"abstract":"<jats:p>\n            Multi-view volumetric rendering techniques have recently shown great potential in modeling and synthesizing high-quality head avatars. A common approach to capture full head dynamic performances is to track the underlying geometry using a mesh-based template or 3D cube-based graphics primitives. While these model-based approaches achieve promising results, they often fail to learn complex geometric details such as the mouth interior, hair, and topological changes over time. This article presents a novel approach to building highly photorealistic digital head avatars. Our method learns a canonical space via an implicit function parameterized by a neural network. It leverages multiresolution hash encoding in the learned feature space, allowing for high quality, faster training, and high-resolution rendering. At test time, our method is driven by a monocular RGB video. Here, an image encoder extracts face-specific features that also condition the learnable canonical space. This encourages deformation-dependent texture variations during training. We also propose a novel optical flow-based loss that ensures correspondences in the learned canonical space, thus encouraging artifact-free and temporally consistent renderings. We show results on challenging facial expressions and show free-viewpoint renderings at interactive real-time rates for a resolution of 480\n            <jats:italic>x<\/jats:italic>\n            270. Our method outperforms related approaches both visually and numerically. We will release our multiple-identity dataset to encourage further research.\n          <\/jats:p>","DOI":"10.1145\/3649889","type":"journal-article","created":{"date-parts":[[2024,2,29]],"date-time":"2024-02-29T12:29:10Z","timestamp":1709209750000},"page":"1-24","update-policy":"https:\/\/doi.org\/10.1145\/crossmark-policy","source":"Crossref","is-referenced-by-count":14,"title":["HQ3DAvatar: High-quality Implicit 3D Head Avatar"],"prefix":"10.1145","volume":"43","author":[{"ORCID":"https:\/\/orcid.org\/0009-0007-6985-7159","authenticated-orcid":false,"given":"Kartik","family":"Teotia","sequence":"first","affiliation":[{"name":"Max Planck Institute for Informatics and Saarland University, Saarbrucken, Germany"}]},{"ORCID":"https:\/\/orcid.org\/0009-0007-5906-8666","authenticated-orcid":false,"given":"Mallikarjun B","family":"R","sequence":"additional","affiliation":[{"name":"Max Planck Institute for Informatics and Saarland University, Saarbrucken, Germany"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5825-9467","authenticated-orcid":false,"given":"Xingang","family":"Pan","sequence":"additional","affiliation":[{"name":"Max Planck Institute for Informatics, Saarbrucken, Germany and Nanyang Technological University, Singapore, Singapore"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-0858-0882","authenticated-orcid":false,"given":"Hyeongwoo","family":"Kim","sequence":"additional","affiliation":[{"name":"Imperial College London, London, United Kingdom"}]},{"ORCID":"https:\/\/orcid.org\/0009-0001-8273-6737","authenticated-orcid":false,"given":"Pablo","family":"Garrido","sequence":"additional","affiliation":[{"name":"Flawless AI, Los Angeles, United States of America"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8727-0895","authenticated-orcid":false,"given":"Mohamed","family":"Elgharib","sequence":"additional","affiliation":[{"name":"Max Planck Institute for Informatics, Saarbrucken, Germany"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6104-6625","authenticated-orcid":false,"given":"Christian","family":"Theobalt","sequence":"additional","affiliation":[{"name":"Max Planck Institute for Informatics and Saarland University, Saarbrucken, Germany"}]}],"member":"320","published-online":{"date-parts":[[2024,4,9]]},"reference":[{"key":"e_1_3_2_2_1","article-title":"3DAvatarGAN: Bridging domains for personalized editable avatars","volume":"2301","author":"Abdal Rameen","year":"2023","unstructured":"Rameen Abdal, Hsin-Ying Lee, Peihao Zhu, Menglei Chai, Aliaksandr Siarohin, Peter Wonka, and Sergey Tulyakov. 2023. 3DAvatarGAN: Bridging domains for personalized editable avatars. CoRR abs\/2301.02700 (2023).","journal-title":"CoRR"},{"key":"e_1_3_2_3_1","unstructured":"Matthew Amodio David van Dijk Ruth Montgomery Guy Wolf and Smita Krishnaswamy. 2019. Out-of-sample Extrapolation with Neuron Editing. arxiv:q-bio.QM\/1805.12198 (2019)."},{"key":"e_1_3_2_4_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01972"},{"key":"e_1_3_2_5_1","article-title":"High-fidelity facial avatar reconstruction from monocular video with generative priors","volume":"2211","author":"Bai Yunpeng","year":"2022","unstructured":"Yunpeng Bai, Yanbo Fan, Xuan Wang, Yong Zhang, Jingxiang Sun, Chun Yuan, and Ying Shan. 2022. High-fidelity facial avatar reconstruction from monocular video with generative priors. CoRR abs\/2211.15064 (2022).","journal-title":"CoRR"},{"key":"e_1_3_2_6_1","volume-title":"Conference on Advances in Neural Information Processing Systems.","author":"Bergman Alexander W.","year":"2022","unstructured":"Alexander W. Bergman, Petr Kellnhofer, Yifan Wang, Eric R. Chan, David B. Lindell, and Gordon Wetzstein. 2022. Generative neural articulated radiance fields. In Conference on Advances in Neural Information Processing Systems."},{"key":"e_1_3_2_7_1","doi-asserted-by":"publisher","DOI":"10.1145\/3618401"},{"key":"e_1_3_2_8_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.00337"},{"key":"e_1_3_2_9_1","doi-asserted-by":"publisher","DOI":"10.1145\/3450626.3459806"},{"key":"e_1_3_2_10_1","doi-asserted-by":"publisher","DOI":"10.1145\/3528223.3530143"},{"key":"e_1_3_2_11_1","doi-asserted-by":"publisher","DOI":"10.1109\/TVCG.2013.249"},{"key":"e_1_3_2_12_1","doi-asserted-by":"publisher","DOI":"10.1145\/2897824.2925873"},{"key":"e_1_3_2_13_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01565"},{"key":"e_1_3_2_14_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.00574"},{"key":"e_1_3_2_15_1","doi-asserted-by":"publisher","DOI":"10.1145\/3478513.3480509"},{"key":"e_1_3_2_16_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.01286"},{"key":"e_1_3_2_17_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01041"},{"key":"e_1_3_2_18_1","doi-asserted-by":"publisher","DOI":"10.1145\/3618399"},{"key":"e_1_3_2_19_1","doi-asserted-by":"publisher","DOI":"10.1145\/3395208"},{"key":"e_1_3_2_20_1","doi-asserted-by":"publisher","DOI":"10.1145\/3414685.3417808"},{"key":"e_1_3_2_21_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.00542"},{"key":"e_1_3_2_22_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.00854"},{"key":"e_1_3_2_23_1","doi-asserted-by":"publisher","DOI":"10.1145\/3550454.3555501"},{"key":"e_1_3_2_24_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00125"},{"key":"e_1_3_2_25_1","first-page":"75","volume-title":"Conference on Automatic Face & Gesture Recognition","author":"Gerig Thomas","year":"2018","unstructured":"Thomas Gerig, Andreas Morel-Forster, Clemens Blumer, Bernhard Egger, Marcel L\u00fcthi, Sandro Sch\u00f6nborn, and Thomas Vetter. 2018. Morphable face models\u2014An open framework. In Conference on Automatic Face & Gesture Recognition. IEEE Computer Society, 75\u201382."},{"key":"e_1_3_2_26_1","volume-title":"International Conference on Learning Representations.","author":"Gu Jiatao","year":"2022","unstructured":"Jiatao Gu, Lingjie Liu, Peng Wang, and Christian Theobalt. 2022. StyleNeRF: A style-based 3D aware generator for high-resolution image synthesis. In International Conference on Learning Representations.OpenReview.net."},{"key":"e_1_3_2_27_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01973"},{"key":"e_1_3_2_28_1","doi-asserted-by":"publisher","DOI":"10.1145\/2766974"},{"key":"e_1_3_2_29_1","doi-asserted-by":"publisher","DOI":"10.1145\/3478513.3480546"},{"key":"e_1_3_2_30_1","doi-asserted-by":"publisher","DOI":"10.1145\/3197517.3201283"},{"key":"e_1_3_2_31_1","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2021.3125598"},{"key":"e_1_3_2_32_1","series-title":"European Conference on Computer Vision.","first-page":"419","volume":"13692","author":"Li Ruilong","year":"2022","unstructured":"Ruilong Li, Julian Tanke, Minh Vo, Michael Zollh\u00f6fer, J\u00fcrgen Gall, Angjoo Kanazawa, and Christoph Lassner. 2022b. TAVA: Template-free animatable volumetric actors. In European Conference on Computer Vision. (Lecture Notes in Computer Science), Vol. 13692. Springer, 419\u2013436."},{"key":"e_1_3_2_33_1","doi-asserted-by":"publisher","DOI":"10.1145\/3130800.3130813"},{"key":"e_1_3_2_34_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.00544"},{"key":"e_1_3_2_35_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.00643"},{"key":"e_1_3_2_36_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.00593"},{"key":"e_1_3_2_37_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.00865"},{"key":"e_1_3_2_38_1","doi-asserted-by":"publisher","DOI":"10.1145\/3197517.3201401"},{"key":"e_1_3_2_39_1","doi-asserted-by":"publisher","DOI":"10.1145\/3306346.3323020"},{"key":"e_1_3_2_40_1","doi-asserted-by":"publisher","DOI":"10.1145\/3450626.3459863"},{"key":"e_1_3_2_41_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.00013"},{"key":"e_1_3_2_42_1","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV48922.2021.01357"},{"key":"e_1_3_2_43_1","unstructured":"Metashape. 2020. Agisoft Metashape (Version 1.8.4) (Software). (2020). Retrieved from https:\/\/www.agisoft.com\/downloads\/installer\/"},{"key":"e_1_3_2_44_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-19784-0_11"},{"key":"e_1_3_2_45_1","doi-asserted-by":"publisher","DOI":"10.1145\/3503250"},{"key":"e_1_3_2_46_1","doi-asserted-by":"publisher","DOI":"10.1145\/3528223.3530127"},{"key":"e_1_3_2_47_1","doi-asserted-by":"publisher","DOI":"10.1145\/3272127.3275075"},{"key":"e_1_3_2_48_1","first-page":"13493","volume-title":"IEEE Conference on Computer Vision and Pattern Recognition.","author":"Or-El Roy","year":"2022","unstructured":"Roy Or-El, Xuan Luo, Mengyi Shan, Eli Shechtman, Jeong Joon Park, and Ira Kemelmacher-Shlizerman. 2022. StyleSDF: High-resolution 3D-consistent image and geometry generation. In IEEE Conference on Computer Vision and Pattern Recognition.IEEE, 13493\u201313503."},{"key":"e_1_3_2_49_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00025"},{"key":"e_1_3_2_50_1","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV48922.2021.00581"},{"key":"e_1_3_2_51_1","doi-asserted-by":"publisher","DOI":"10.1145\/3478513.3480487"},{"key":"e_1_3_2_52_1","doi-asserted-by":"publisher","DOI":"10.5244\/C.29.41"},{"key":"e_1_3_2_53_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.01156"},{"key":"e_1_3_2_54_1","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV48922.2021.00557"},{"key":"e_1_3_2_55_1","article-title":"Facial geometric detail recovery via implicit representation","volume":"2203","author":"Ren Xingyu","year":"2022","unstructured":"Xingyu Ren, Alexandros Lattas, Baris Gecer, Jiankang Deng, Chao Ma, Xiaokang Yang, and Stefanos Zafeiriou. 2022. Facial geometric detail recovery via implicit representation. CoRR abs\/2203.09692 (2022).","journal-title":"CoRR"},{"key":"e_1_3_2_56_1","article-title":"Synthesizing facial photometries and corresponding geometries using generative adversarial networks","volume":"1901","author":"Shamai Gil","year":"2019","unstructured":"Gil Shamai, Ron Slossberg, and Ron Kimmel. 2019. Synthesizing facial photometries and corresponding geometries using generative adversarial networks. CoRR abs\/1901.06551 (2019).","journal-title":"CoRR"},{"key":"e_1_3_2_57_1","article-title":"Controllable 3D face synthesis with conditional generative occupancy fields","volume":"2206","author":"Sun Keqiang","year":"2022","unstructured":"Keqiang Sun, Shangzhe Wu, Zhaoyang Huang, Ning Zhang, Quan Wang, and Hongsheng Li. 2022. Controllable 3D face synthesis with conditional generative occupancy fields. CoRR abs\/2206.08361 (2022).","journal-title":"CoRR"},{"key":"e_1_3_2_58_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.01120"},{"key":"e_1_3_2_59_1","unstructured":"Jiaxiang Tang. 2022. Torch-ngp: A PyTorch Implementation of Instant-ngp. (2022). Retrieved from https:\/\/github.com\/ashawkey\/torch-ngp"},{"key":"e_1_3_2_60_1","doi-asserted-by":"publisher","DOI":"10.1145\/3414685.3417803"},{"key":"e_1_3_2_61_1","doi-asserted-by":"publisher","DOI":"10.1111\/cgf.14507"},{"key":"e_1_3_2_62_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00270"},{"key":"e_1_3_2_63_1","doi-asserted-by":"publisher","DOI":"10.1145\/3306346.3323035"},{"key":"e_1_3_2_64_1","doi-asserted-by":"publisher","DOI":"10.1145\/2929464.2929475"},{"key":"e_1_3_2_65_1","doi-asserted-by":"publisher","DOI":"10.1145\/3292039"},{"key":"e_1_3_2_66_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00122"},{"key":"e_1_3_2_67_1","first-page":"55:1\u201355:9","volume-title":"SIGGRAPH","author":"Wang Daoye","year":"2022","unstructured":"Daoye Wang, Prashanth Chandran, Gaspard Zoss, Derek Bradley, and Paulo F. U. Gotardo. 2022a. MoRF: Morphable radiance fields for multiview neural head modeling. In SIGGRAPH. ACM, 55:1\u201355:9."},{"key":"e_1_3_2_68_1","doi-asserted-by":"publisher","DOI":"10.1145\/3588432.3591517"},{"key":"e_1_3_2_69_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.00991"},{"key":"e_1_3_2_70_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.00565"},{"key":"e_1_3_2_71_1","doi-asserted-by":"publisher","DOI":"10.1109\/TIP.2003.819861"},{"key":"e_1_3_2_72_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.00605"},{"key":"e_1_3_2_73_1","article-title":"GRAM-HD: 3D-consistent image generation at high resolution with generative radiance manifolds","volume":"2206","author":"Xiang Jianfeng","year":"2022","unstructured":"Jianfeng Xiang, Jiaolong Yang, Yu Deng, and Xin Tong. 2022. GRAM-HD: 3D-consistent image generation at high resolution with generative radiance manifolds. CoRR abs\/2206.07255 (2022).","journal-title":"CoRR"},{"key":"e_1_3_2_74_1","doi-asserted-by":"publisher","DOI":"10.1145\/3588432.3591567"},{"key":"e_1_3_2_75_1","doi-asserted-by":"publisher","DOI":"10.1145\/3197517.3201364"},{"key":"e_1_3_2_76_1","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV48922.2021.00570"},{"key":"e_1_3_2_77_1","article-title":"FLNeRF: 3D facial landmarks estimation in neural radiance fields","volume":"2211","author":"Zhang Hao","year":"2022","unstructured":"Hao Zhang, Tianyuan Dai, Yu-Wing Tai, and Chi-Keung Tang. 2022. FLNeRF: 3D facial landmarks estimation in neural radiance fields. CoRR abs\/2211.11202 (2022).","journal-title":"CoRR"},{"key":"e_1_3_2_78_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00068"},{"key":"e_1_3_2_79_1","doi-asserted-by":"publisher","DOI":"10.1145\/3626316"},{"key":"e_1_3_2_80_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01318"},{"key":"e_1_3_2_81_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.02017"},{"key":"e_1_3_2_82_1","article-title":"Towards metrical reconstruction of human faces","author":"Zielonka Wojciech","year":"2022","unstructured":"Wojciech Zielonka, Timo Bolkart, and Justus Thies. 2022. Towards metrical reconstruction of human faces. In European Conference on Computer Vision.","journal-title":"European Conference on Computer Vision"},{"key":"e_1_3_2_83_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.00444"},{"key":"e_1_3_2_84_1","doi-asserted-by":"publisher","DOI":"10.1111\/cgf.13382"}],"container-title":["ACM Transactions on Graphics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3649889","content-type":"unspecified","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3649889","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,6,18]],"date-time":"2025-06-18T22:54:07Z","timestamp":1750287247000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3649889"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,4,9]]},"references-count":83,"journal-issue":{"issue":"3","published-print":{"date-parts":[[2024,6,30]]}},"alternative-id":["10.1145\/3649889"],"URL":"https:\/\/doi.org\/10.1145\/3649889","relation":{},"ISSN":["0730-0301","1557-7368"],"issn-type":[{"value":"0730-0301","type":"print"},{"value":"1557-7368","type":"electronic"}],"subject":[],"published":{"date-parts":[[2024,4,9]]},"assertion":[{"value":"2023-03-15","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2024-02-13","order":1,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2024-04-09","order":2,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}