{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,1]],"date-time":"2026-05-01T16:52:31Z","timestamp":1777654351191,"version":"3.51.4"},"reference-count":62,"publisher":"IEEE","license":[{"start":{"date-parts":[[2023,5,29]],"date-time":"2023-05-29T00:00:00Z","timestamp":1685318400000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2023,5,29]],"date-time":"2023-05-29T00:00:00Z","timestamp":1685318400000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"DOI":"10.13039\/100000006","name":"ONR","doi-asserted-by":"publisher","id":[{"id":"10.13039\/100000006","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100014036","name":"MURI","doi-asserted-by":"publisher","award":["N00014-22-1-2740"],"award-info":[{"award-number":["N00014-22-1-2740"]}],"id":[{"id":"10.13039\/100014036","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100000001","name":"NSF","doi-asserted-by":"publisher","award":["2120095"],"award-info":[{"award-number":["2120095"]}],"id":[{"id":"10.13039\/100000001","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100016443","name":"Amazon","doi-asserted-by":"publisher","id":[{"id":"10.13039\/100016443","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100012884","name":"Bosch","doi-asserted-by":"publisher","id":[{"id":"10.13039\/100012884","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100011800","name":"Salesforce","doi-asserted-by":"publisher","id":[{"id":"10.13039\/100011800","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2023,5,29]]},"DOI":"10.1109\/icra48891.2023.10160461","type":"proceedings-article","created":{"date-parts":[[2023,7,4]],"date-time":"2023-07-04T17:20:56Z","timestamp":1688491256000},"page":"704-711","source":"Crossref","is-referenced-by-count":10,"title":["Sonicverse: A Multisensory Simulation Platform for Embodied Household Agents that See and Hear"],"prefix":"10.1109","author":[{"given":"Ruohan","family":"Gao","sequence":"first","affiliation":[{"name":"Stanford University"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Hao","family":"Li","sequence":"additional","affiliation":[{"name":"Stanford University"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Gokul","family":"Dharan","sequence":"additional","affiliation":[{"name":"Stanford University"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Zhuzhu","family":"Wang","sequence":"additional","affiliation":[{"name":"Stanford University"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Chengshu","family":"Li","sequence":"additional","affiliation":[{"name":"Stanford University"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Fei","family":"Xia","sequence":"additional","affiliation":[{"name":"Stanford University"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Silvio","family":"Savarese","sequence":"additional","affiliation":[{"name":"Stanford University"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Li","family":"Fei-Fei","sequence":"additional","affiliation":[{"name":"Stanford University"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jiajun","family":"Wu","sequence":"additional","affiliation":[{"name":"Stanford University"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref13","article-title":"Threed-world: A platform for interactive multi-modal physical simulation","author":"gan","year":"2021","journal-title":"NeurIPS Datasets and Benchmarks Track"},{"key":"ref57","article-title":"Learning exploration policies for navigation","author":"chen","year":"2019","journal-title":"ArXiv Preprint"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00886"},{"key":"ref56","article-title":"U-net: Convolutional networks for biomedical image segmentation","author":"ronneberger","year":"0","journal-title":"International Conference on Medical Image Computing and Computer-Assisted Intervention"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1109\/LRA.2020.2974707"},{"key":"ref59","doi-asserted-by":"publisher","DOI":"10.1177\/0278364916684382"},{"key":"ref14","article-title":"robosuite: A modular simulation framework and benchmark for robot learning","author":"zhu","year":"2020","journal-title":"ArXiv Preprint"},{"key":"ref58","doi-asserted-by":"publisher","DOI":"10.1177\/0278364911421039"},{"key":"ref53","year":"0","journal-title":"OpenVRML"},{"key":"ref52","doi-asserted-by":"publisher","DOI":"10.1109\/3DV.2017.00081"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.00323"},{"key":"ref55","article-title":"Occupancy anticipation for efficient exploration and navigation","author":"ramakrishnan","year":"0","journal-title":"ECCV"},{"key":"ref10","article-title":"igibson, a simulation environment for interactive tasks in large realisticscenes","author":"shen","year":"0","journal-title":"IROS"},{"key":"ref54","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00063"},{"key":"ref17","article-title":"igibson 2.0: Object-centric simulation for robot learning of everyday household tasks","author":"li","year":"0","journal-title":"CoRL"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-58539-6_2"},{"key":"ref19","article-title":"Learning to set waypoints for audio-visual navigation","author":"chen","year":"0","journal-title":"ICLRE"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1109\/ICRA40945.2020.9197008"},{"key":"ref51","article-title":"Efficient encoding and decoding of binaural sound with resonance audio","author":"gorzel","year":"0","journal-title":"in AES International Conference on Immersive and Interactive Audio"},{"key":"ref50","article-title":"Geometry-aware multi-task learning for binaural audio generation from video","author":"garg","year":"0","journal-title":"BMVC"},{"key":"ref46","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-01216-8_16"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00458"},{"key":"ref48","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00041"},{"key":"ref47","article-title":"Discriminative sounding objects localization via self-supervised audiovisual matching","author":"hu","year":"0","journal-title":"NeurIPS"},{"key":"ref42","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.01049"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-01246-5_35"},{"key":"ref44","article-title":"Into the wild with audioscope: Unsupervised audio-visual separation of on-screen sounds","author":"tzinis","year":"0","journal-title":"ICLRE"},{"key":"ref43","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00398"},{"key":"ref49","article-title":"Self-supervised generation of spatial audio for 360° video","author":"morgado","year":"0","journal-title":"NeurIPS"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.3390\/s19010176"},{"key":"ref7","article-title":"Semi-parametric topological memory for navigation","author":"savinov","year":"0","journal-title":"ICLRE"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.01111"},{"key":"ref4","article-title":"Learning to explore using active neural slam","author":"chaplot","year":"0","journal-title":"ICLRE"},{"key":"ref3","article-title":"Dd-ppo: Learning near-perfect pointgoal navigators from 2.5 billion frames","author":"wijmans","year":"0","journal-title":"ICLRE"},{"key":"ref6","article-title":"Object goal navigation using goal-oriented semantic exploration","author":"chaplot","year":"0","journal-title":"NeurIPS"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2017.60"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-01219-9_3"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01034"},{"key":"ref34","article-title":"Objectfolder: A dataset of objects with implicit visual, auditory, and tactile representations","author":"gao","year":"0","journal-title":"CoRL"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2017.73"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-46448-0_48"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1162\/neco.1997.9.8.1735"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.01112"},{"key":"ref33","article-title":"Noisy agents: Self-supervised exploration by predicting auditory events","author":"gan","year":"2020","journal-title":"ArXiv Preprint"},{"key":"ref32","article-title":"See, hear, explore: Curiosity via audio-visual association","author":"dean","year":"0","journal-title":"NeurIPS"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00943"},{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.769"},{"key":"ref39","article-title":"Co-training of audio and video representations from self-supervised temporal synchronization","author":"korbar","year":"0","journal-title":"NeurIPS"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-01231-1_39"},{"key":"ref24","article-title":"Ai2-thor: An interactive 3d environment for visual ai","author":"kolve","year":"2017","journal-title":"ArXiv Preprint"},{"key":"ref23","article-title":"Habitat 2.0: Training home assistants to rearrange their habitat","author":"szot","year":"0","journal-title":"NeurIPS"},{"key":"ref26","article-title":"Soundspaces 2.0: A simulation platform for visual-acoustic learning","author":"chen","year":"0","journal-title":"ArXiv"},{"key":"ref25","article-title":"Meta-world: A benchmark and evaluation for multi-task and meta reinforcement learning","author":"yu","year":"0","journal-title":"CoRL"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.01526"},{"key":"ref22","article-title":"Visuale-choes: Spatial image representation learning through echolocation","author":"gao","year":"0","journal-title":"ECCV"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV48922.2021.00122"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2021.eacl-main.111"},{"key":"ref27","article-title":"Objectnav revisited: On evaluation of embodied agents navigating to objects","author":"batra","year":"2020","journal-title":"ArXiv Preprint"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00387"},{"key":"ref60","doi-asserted-by":"publisher","DOI":"10.1109\/ICRA.2019.8793500"},{"key":"ref62","article-title":"On evaluation of embodied navigation agents","author":"anderson","year":"2018","journal-title":"ArXiv Preprint"},{"key":"ref61","doi-asserted-by":"publisher","DOI":"10.1109\/ICRA.2019.8793769"}],"event":{"name":"2023 IEEE International Conference on Robotics and Automation (ICRA)","location":"London, United Kingdom","start":{"date-parts":[[2023,5,29]]},"end":{"date-parts":[[2023,6,2]]}},"container-title":["2023 IEEE International Conference on Robotics and Automation (ICRA)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/10160211\/10160212\/10160461.pdf?arnumber=10160461","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,7,24]],"date-time":"2023-07-24T17:31:08Z","timestamp":1690219868000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/10160461\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,5,29]]},"references-count":62,"URL":"https:\/\/doi.org\/10.1109\/icra48891.2023.10160461","relation":{},"subject":[],"published":{"date-parts":[[2023,5,29]]}}}