{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,15]],"date-time":"2026-07-15T17:03:25Z","timestamp":1784135005899,"version":"3.55.0"},"reference-count":45,"publisher":"Acoustical Society of America (ASA)","issue":"1","content-domain":{"domain":["pubs.aip.org"],"crossmark-restriction":true},"short-container-title":[],"published-print":{"date-parts":[[2026,7,1]]},"abstract":"<jats:p>Sparse room impulse responses (RIRs) encode indoor spatial structure through direct sound, early reflections, and reverberation shaped by room boundaries and objects. However, estimating a spatial layout from only a few source\u2013receiver RIR measurements is highly ill-posed because the acoustic field is spatially undersampled and different layouts may produce similar echo patterns. This problem is important for acoustic sensing, robotic navigation, and environmental perception when dense measurements or reliable visual observations are unavailable. We propose a two-stage framework for spatial layout inference from sparse RIRs. First, an implicit neural acoustic field reconstructs a continuous acoustic representation from sparse source\u2013receiver responses. Acoustic reciprocity is introduced as an unsupervised physical regularizer to improve interpolation in unobserved regions, and the reconstructed field is represented in the Mel-spectrogram domain as a compact intermediate feature. Second, a cross-modal acoustic-to-visual network translates this reconstructed field into an image-based spatial layout proxy. Experiments on high-fidelity simulations and a public indoor acoustic benchmark show improved acoustic field reconstruction and reliable spatial layout inference from sparse RIR measurements.<\/jats:p>","DOI":"10.1121\/10.0044365","type":"journal-article","created":{"date-parts":[[2026,7,15]],"date-time":"2026-07-15T16:03:46Z","timestamp":1784131426000},"page":"493-506","update-policy":"https:\/\/doi.org\/10.1063\/aip-crossmark-policy-page","source":"Crossref","is-referenced-by-count":0,"title":["Physics-regularized neural acoustic fields for spatial layout inference from sparse room impulse responses"],"prefix":"10.1121","volume":"160","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-6367-7583","authenticated-orcid":false,"given":"Haoran","family":"Wang","sequence":"first","affiliation":[{"name":"School of Electronic Science and Engineering, Nanjing University , Nanjing, Jiangsu 210023,","place":["China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Shengzhi","family":"Qiang","sequence":"additional","affiliation":[{"name":"School of Electronic Science and Engineering, Nanjing University , Nanjing, Jiangsu 210023,","place":["China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Qi","family":"Wang","sequence":"additional","affiliation":[{"name":"School of Electronic Science and Engineering, Nanjing University , Nanjing, Jiangsu 210023,","place":["China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Qing","family":"Lu","sequence":"additional","affiliation":[{"name":"School of Electronic Science and Engineering, Nanjing University , Nanjing, Jiangsu 210023,","place":["China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6316-9426","authenticated-orcid":false,"given":"Yuanqing","family":"Wang","sequence":"additional","affiliation":[{"name":"School of Electronic Science and Engineering, Nanjing University , Nanjing, Jiangsu 210023,","place":["China"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"231","published-online":{"date-parts":[[2026,7,15]]},"reference":[{"issue":"10","key":"2026071512034190400_c1","doi-asserted-by":"publisher","first-page":"2683","DOI":"10.1109\/TASL.2012.2210877","article-title":"Inference of room geometry from acoustic impulse responses","volume":"20","year":"2012","journal-title":"IEEE Trans. Audio. Speech Lang. Process."},{"key":"2026071512034190400_c2","first-page":"1581","article-title":"Batvision: Learning to see 3D spatial layout with two ears","year":"2020"},{"key":"2026071512034190400_c3","doi-asserted-by":"publisher","first-page":"581","DOI":"10.1121\/1.4940668","article-title":"Modeling human echolocation of near-range targets with an audible sonar","volume":"139","year":"2016","journal-title":"J. Acoust. Soc. Am."},{"issue":"6","key":"2026071512034190400_c4","doi-asserted-by":"publisher","first-page":"1526","DOI":"10.1109\/TBME.2015.2393371","article-title":"A device for human ultrasonic echolocation","volume":"62","year":"2015","journal-title":"IEEE Trans. Biomed. Eng."},{"key":"2026071512034190400_c5","first-page":"5140","article-title":"Automatic color correction for 3D reconstruction of underwater scenes"},{"key":"2026071512034190400_c6","doi-asserted-by":"publisher","first-page":"1833","DOI":"10.1007\/s10462-022-10210-8","article-title":"Overview of behavior recognition based on deep learning","volume":"56","year":"2023","journal-title":"Artif. Intell. Rev."},{"key":"2026071512034190400_c7","first-page":"I","article-title":"Clear underwater vision","year":"2004"},{"key":"2026071512034190400_c8","article-title":"Underwater photogrammetry in very shallow waters: Main challenges and caustics effect removal","year":"2018"},{"key":"2026071512034190400_c9","doi-asserted-by":"publisher","first-page":"112341","DOI":"10.1016\/j.jcp.2023.112341","article-title":"A neural network warm-start approach for the inverse acoustic obstacle scattering problem","volume":"490","year":"2023","journal-title":"J. Comput. Phys."},{"key":"2026071512034190400_c10","first-page":"1","article-title":"Operating frequency trade-offs in the design of high resolution synthetic aperture sonars for small platforms","year":"2014"},{"key":"2026071512034190400_c11","doi-asserted-by":"publisher","first-page":"125522","DOI":"10.1109\/ACCESS.2019.2939005","article-title":"Accurate underwater ATR in forward-looking sonar imagery using deep convolutional neural networks","volume":"7","year":"2019","journal-title":"IEEE Access."},{"key":"2026071512034190400_c12","doi-asserted-by":"publisher","first-page":"297","DOI":"10.1109\/JOE.2012.2226643","article-title":"Improving sonar performance in shallow water using adaptive beamforming","volume":"38","year":"2013","journal-title":"IEEE J. Oceanic Eng."},{"key":"2026071512034190400_c13","first-page":"149","article-title":"Underwater acoustics","volume-title":"Springer Handbook of Acoustics","year":"2014"},{"key":"2026071512034190400_c14","first-page":"3","article-title":"Multipath propagation and its effects on sonar design and performance in the real ocean","year":"1976"},{"key":"2026071512034190400_c15","doi-asserted-by":"publisher","first-page":"44","DOI":"10.1109\/JOE.2002.808203","article-title":"Adaptive beamforming of a towed array during a turn","volume":"28","year":"2003","journal-title":"IEEE J. Oceanic Eng."},{"key":"2026071512034190400_c16","article-title":"Neural synthesis of binaural speech from mono audio","year":"2021"},{"key":"2026071512034190400_c17","first-page":"23689","article-title":"BinauralGrad: A two-stage conditional diffusion probabilistic model for binaural audio synthesis","year":"2022"},{"key":"2026071512034190400_c18","first-page":"1","article-title":"Neural Fourier shift for binaural speech rendering","year":"2023"},{"key":"2026071512034190400_c19","first-page":"1","article-title":"Dual position attention time-frequency network for binaural audio synthesis","year":"2025"},{"issue":"6","key":"2026071512034190400_c20","doi-asserted-by":"publisher","first-page":"1790","DOI":"10.3390\/s25061790","article-title":"Two-stage Unet with gated-conv fusion for binaural audio synthesis","volume":"25","year":"2025","journal-title":"Sensors"},{"key":"2026071512034190400_c21","first-page":"7462","article-title":"Implicit neural representations with periodic activation functions","year":"2020"},{"key":"2026071512034190400_c22","doi-asserted-by":"publisher","first-page":"3165","DOI":"10.48550\/arXiv.2204.00628","article-title":"Learning neural acoustic fields","volume":"35","year":"2022","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"2026071512034190400_c23","first-page":"37472","article-title":"AV-NeRF: Learning neural fields for realworld audio-visual scene synthesis","year":"2023"},{"key":"2026071512034190400_c24","article-title":"NeRAF: 3D scene infused neural radiance and acoustic fields","year":"2025"},{"key":"2026071512034190400_c25","first-page":"8144","article-title":"INRAS: Implicit neural representation for audio scenes","year":"2022"},{"key":"2026071512034190400_c26","first-page":"924","article-title":"Mesh2ir: Neural acoustic impulse response generator for complex 3D scenes","year":"2022"},{"key":"2026071512034190400_c27","first-page":"2522","article-title":"Few-shot audio-visual learning of environment acoustics","year":"2022"},{"issue":"2","key":"2026071512034190400_c28","doi-asserted-by":"publisher","first-page":"1048","DOI":"10.1121\/10.0024750","article-title":"Room impulse response reconstruction with physics-informed deep learning","volume":"155","year":"2024","journal-title":"J. Acoust. Soc. Am."},{"issue":"2","key":"2026071512034190400_c29","doi-asserted-by":"publisher","first-page":"e2312159120","DOI":"10.1073\/pnas.2312159120","article-title":"Sound propagation in realistic interactive 3D scenes with parameterized sources using deep neural operators","volume":"121","year":"2024","journal-title":"Proc. Natl. Acad. Sci. U.S.A."},{"key":"2026071512034190400_c30","doi-asserted-by":"publisher","first-page":"436","DOI":"10.1109\/TASLP.2020.3043115","article-title":"Room acoustical parameter estimation from room impulse responses using deep neural networks","volume":"29","year":"2020","journal-title":"IEEE\/ACM Trans. Audio. Speech Lang. Process."},{"key":"2026071512034190400_c31","unstructured":"C.\n            Tuna\n           and E. A.Habets, \u201cData-driven 3D room geometry inference with a linear loudspeaker array and a single microphone,\u201d arXiv:2308.14611 (2023)."},{"key":"2026071512034190400_c32","doi-asserted-by":"publisher","first-page":"4768","DOI":"10.1109\/TASLP.2024.3485516","article-title":"EchoScan: Scanning complex room geometries via acoustic echoes","volume":"32","year":"2024","journal-title":"IEEE\/ACM Trans. Audio. Speech Lang. Process."},{"issue":"8","key":"2026071512034190400_c33","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1145\/3736768","article-title":"Echo depth estimation via attention-based hierarchical multi-scale feature fusion network","volume":"21","year":"2025","journal-title":"ACM Trans. Multimedia Comput. Commun. Appl."},{"key":"2026071512034190400_c34","article-title":"PANO-ECHO: PANOramic depth prediction enhancement with ECHO features","year":"2024"},{"key":"2026071512034190400_c35","first-page":"8268","article-title":"Beyond image to depth: Improving depth prediction using echoes","year":"2021"},{"key":"2026071512034190400_c36","first-page":"658","article-title":"Visual-echoes: Spatial image representation learning through echolocation","year":"2020"},{"key":"2026071512034190400_c37","unstructured":"L.\n            Zhu\n          , E.Rahtu, and H.Zhao, \u201cBeyond visual field of view: Perceiving 3D environment with echoes and vision,\u201d arXiv:2207.01136 (2022)."},{"key":"2026071512034190400_c38","first-page":"1125","article-title":"Image-to-image translation with conditional adversarial networks","year":"2017"},{"key":"2026071512034190400_c39","first-page":"2813","article-title":"Least squares generative adversarial networks","year":"2017"},{"key":"2026071512034190400_c40","first-page":"17","article-title":"SoundSpaces: Audio-visual navigation in 3D environments","year":"2020"},{"key":"2026071512034190400_c41","doi-asserted-by":"crossref","unstructured":"A.\n            Chang\n          , A.Dai, T.Funkhouser, M.Halber, M.Niessner, M.Savva, and Y.Zhang, \u201cMatterport3d: Learning from RGB-D data in indoor environments,\u201d arXiv:1709.06158 (2017).","DOI":"10.1109\/3DV.2017.00081"},{"key":"2026071512034190400_c42","unstructured":"J.\n            Straub\n          , T.Whelan, L.Ma, Y.Chen, E.Wijmans, S.Green, and J. J.Engel, \u201cThe replica dataset: A digital replica of indoor spaces,\u201d arXiv:1906.05797 (2019)."},{"key":"2026071512034190400_c43","unstructured":"D.\n            Kingma\n           and J.Ba, \u201cAdam: A method for stochastic optimization,\u201d arXiv:1412.6980 (2014)."},{"key":"2026071512034190400_c44","first-page":"234","article-title":"U-net: Convolutional networks for biomedical image segmentation","year":"2015"},{"key":"2026071512034190400_c45","first-page":"44600","article-title":"Acoustic volume rendering for neural impulse response fields","year":"2024"}],"container-title":["The Journal of the Acoustical Society of America"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/pubs.aip.org\/asa\/jasa\/article-pdf\/160\/1\/493\/21076780\/493_1_10.0044365.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/pubs.aip.org\/asa\/jasa\/article-pdf\/160\/1\/493\/21076780\/493_1_10.0044365.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,15]],"date-time":"2026-07-15T16:03:47Z","timestamp":1784131427000},"score":1,"resource":{"primary":{"URL":"https:\/\/pubs.aip.org\/jasa\/article\/160\/1\/493\/3398560\/Physics-regularized-neural-acoustic-fields-for"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,7,1]]},"references-count":45,"journal-issue":{"issue":"1","published-print":{"date-parts":[[2026,7,1]]}},"URL":"https:\/\/doi.org\/10.1121\/10.0044365","relation":{},"ISSN":["1520-8524"],"issn-type":[{"value":"1520-8524","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2026,7]]},"published":{"date-parts":[[2026,7,1]]}}}