-
Notifications
You must be signed in to change notification settings - Fork 0
Expand file tree
/
Copy path404.html
More file actions
1 lines (1 loc) · 22.1 KB
/
Copy path404.html
File metadata and controls
1 lines (1 loc) · 22.1 KB
1
<!DOCTYPE html><html lang="en" class="silkscreen_5c214b00-module__BDnqEG__variable plexmono_301f1454-module__3YJGwG__variable"><head><meta charSet="utf-8"/><meta name="viewport" content="width=device-width, initial-scale=1"/><link rel="stylesheet" href="./_next/static/chunks/1zf8qs8h9vuqk.css" data-precedence="next"/><link rel="preload" as="script" fetchPriority="low" href="./_next/static/chunks/0ba1ebj8wcm3s.js"/><script src="./_next/static/chunks/2turxypa9qaj4.js" async=""></script><script src="./_next/static/chunks/01yyx2dlde7l1.js" async=""></script><script src="./_next/static/chunks/turbopack-063p7s1qohf7o.js" async=""></script><script src="./_next/static/chunks/3d-g45tvvtwds.js" async=""></script><title>404: This page could not be found.</title><meta name="robots" content="noindex"/><meta name="next-size-adjust" content=""/><title>SceneAgent | Harvard Computational Robotics Group</title><meta name="description" content="SceneAgent converts 3D captures of real-world scenes into simulation environments with predictive physics for robot policy evaluation, planning and training."/><link rel="author" href="https://computationalrobotics.seas.harvard.edu/"/><meta name="author" content="Luke Hollis"/><link rel="author" href="https://computationalrobotics.seas.harvard.edu/"/><meta name="author" content="Tianxing Fan"/><link rel="author" href="https://computationalrobotics.seas.harvard.edu/"/><meta name="author" content="Heng Yang"/><meta name="keywords" content="SceneAgent,real-to-sim,sim-to-real,3D Gaussian Splatting,simulation environments,predictive physics,robot policy evaluation,policy training,online planning,object segmentation,articulation,deformable objects,digital sisters,robot learning,Harvard Computational Robotics Group"/><meta name="creator" content="Harvard Computational Robotics Group"/><meta name="publisher" content="Harvard Computational Robotics Group"/><meta name="robots" content="index, follow"/><meta name="googlebot" content="index, follow, max-video-preview:-1, max-image-preview:large, max-snippet:-1"/><meta name="citation_title" content="SceneAgent: 3D Capture-Derived Scenes with Predictive Physics for Policy Evaluation and Training Environments"/><meta name="citation_author" content="Luke Hollis"/><meta name="citation_author" content="Tianxing Fan"/><meta name="citation_author" content="Heng Yang"/><meta name="citation_publication_date" content="2026"/><meta name="citation_publisher" content="Harvard Computational Robotics Group"/><meta name="citation_abstract_html_url" content="https://computationalrobotics.seas.harvard.edu/SceneAgent/"/><link rel="canonical" href="https://computationalrobotics.seas.harvard.edu/SceneAgent/"/><meta property="og:title" content="SceneAgent: 3D captures to simulation environments with predictive physics"/><meta property="og:description" content="An agentic pipeline from 3D Gaussian Splatting, photogrammetry and LiDAR captures of real scenes to simulation environments with predictive physics. Harvard Computational Robotics Group, 2026 preprint."/><meta property="og:url" content="https://computationalrobotics.seas.harvard.edu/SceneAgent/"/><meta property="og:site_name" content="Harvard Computational Robotics Group"/><meta property="og:locale" content="en_US"/><meta property="og:image" content="https://computationalrobotics.seas.harvard.edu/SceneAgent/branding/sceneagent-social-share.jpg"/><meta property="og:image:secure_url" content="https://computationalrobotics.seas.harvard.edu/SceneAgent/branding/sceneagent-social-share.jpg"/><meta property="og:image:type" content="image/jpeg"/><meta property="og:image:width" content="1200"/><meta property="og:image:height" content="630"/><meta property="og:image:alt" content="SceneAgent: two Franka robot arms over a simulated tabletop scene, above the paper title and subtitle"/><meta property="og:type" content="website"/><meta name="twitter:card" content="summary_large_image"/><meta name="twitter:title" content="SceneAgent: 3D captures to simulation environments with predictive physics"/><meta name="twitter:description" content="An agentic pipeline from 3D Gaussian Splatting, photogrammetry and LiDAR captures of real scenes to simulation environments with predictive physics. Harvard Computational Robotics Group, 2026 preprint."/><meta name="twitter:image" content="https://computationalrobotics.seas.harvard.edu/SceneAgent/branding/sceneagent-social-share.jpg"/><meta name="twitter:image:alt" content="SceneAgent: two Franka robot arms over a simulated tabletop scene, above the paper title and subtitle"/><link rel="icon" href="./favicon.png" type="image/png"/><script src="./_next/static/chunks/0cz1d0mv5g_q7.js" noModule=""></script></head><body><div hidden=""><!--$--><!--/$--></div><div style="font-family:system-ui,"Segoe UI",Roboto,Helvetica,Arial,sans-serif,"Apple Color Emoji","Segoe UI Emoji";height:100vh;text-align:center;display:flex;flex-direction:column;align-items:center;justify-content:center"><div><style>body{color:#000;background:#fff;margin:0}.next-error-h1{border-right:1px solid rgba(0,0,0,.3)}@media (prefers-color-scheme:dark){body{color:#fff;background:#000}.next-error-h1{border-right:1px solid rgba(255,255,255,.3)}}</style><h1 class="next-error-h1" style="display:inline-block;margin:0 20px 0 0;padding:0 23px 0 0;font-size:24px;font-weight:500;vertical-align:top;line-height:49px">404</h1><div style="display:inline-block"><h2 style="font-size:14px;font-weight:400;line-height:49px;margin:0">This page could not be found.</h2></div></div></div><!--$--><!--/$--><script type="application/ld+json">{"@context":"https://schema.org","@type":"ScholarlyArticle","headline":"SceneAgent: 3D Capture-Derived Scenes with Predictive Physics for Policy Evaluation and Training Environments","name":"SceneAgent","description":"SceneAgent converts 3D captures of real-world scenes into simulation environments with predictive physics for robot policy evaluation, planning and training.","abstract":"SceneAgent converts 3d captures of real-world scenes into simulation environments that can be used for policy evaluation, online planning, and policy training or finetuning. The pipeline primarily focuses on converting environments captured with 3d Gaussian Splatting (3DGS) but also supports scenes captured with photogrammetry and LiDAR. It incorporates 3d semantic features, object segmentation, predictive per-gaussian physics properties, object decomposition, deformability, and articulation. To increase policy generalization, it is also able to generate “digital sisters” or similar versions of each object within an environment. We are still evaluating how well policies trained in this environment perform in the real world, and initial results show promising success rates similar to or improving on other recent papers such as SimFoundry and PolaRiS.","author":[{"@type":"Person","name":"Luke Hollis","affiliation":{"@type":"Organization","name":"Harvard Computational Robotics Group, School of Engineering and Applied Sciences, Harvard University","url":"https://computationalrobotics.seas.harvard.edu/"}},{"@type":"Person","name":"Tianxing Fan","affiliation":{"@type":"Organization","name":"Harvard Computational Robotics Group, School of Engineering and Applied Sciences, Harvard University","url":"https://computationalrobotics.seas.harvard.edu/"}},{"@type":"Person","name":"Heng Yang","affiliation":{"@type":"Organization","name":"Harvard Computational Robotics Group, School of Engineering and Applied Sciences, Harvard University","url":"https://computationalrobotics.seas.harvard.edu/"}}],"publisher":{"@type":"Organization","name":"Harvard Computational Robotics Group","url":"https://computationalrobotics.seas.harvard.edu/"},"datePublished":"2026","creativeWorkStatus":"Preprint","image":"https://computationalrobotics.seas.harvard.edu/SceneAgent/branding/sceneagent-social-share.jpg","url":"https://computationalrobotics.seas.harvard.edu/SceneAgent/","mainEntityOfPage":"https://computationalrobotics.seas.harvard.edu/SceneAgent/","inLanguage":"en","keywords":"SceneAgent, real-to-sim, sim-to-real, 3D Gaussian Splatting, simulation environments, predictive physics, robot policy evaluation, policy training, online planning, object segmentation, articulation, deformable objects, digital sisters, robot learning, Harvard Computational Robotics Group"}</script><script src="./_next/static/chunks/0ba1ebj8wcm3s.js" id="_R_" async=""></script><script>(self.__next_f=self.__next_f||[]).push([0])</script><script>self.__next_f.push([1,"1:\"$Sreact.fragment\"\n2:I[28422,[\"./_next/static/chunks/3d-g45tvvtwds.js\"],\"default\"]\n3:I[4703,[\"./_next/static/chunks/3d-g45tvvtwds.js\"],\"default\"]\n9:I[76400,[\"./_next/static/chunks/3d-g45tvvtwds.js\"],\"default\",1]\n:HL[\"./_next/static/chunks/1zf8qs8h9vuqk.css\",\"style\"]\n4:Tad0,"])</script><script>self.__next_f.push([1,"{\"@context\":\"https://schema.org\",\"@type\":\"ScholarlyArticle\",\"headline\":\"SceneAgent: 3D Capture-Derived Scenes with Predictive Physics for Policy Evaluation and Training Environments\",\"name\":\"SceneAgent\",\"description\":\"SceneAgent converts 3D captures of real-world scenes into simulation environments with predictive physics for robot policy evaluation, planning and training.\",\"abstract\":\"SceneAgent converts 3d captures of real-world scenes into simulation environments that can be used for policy evaluation, online planning, and policy training or finetuning. The pipeline primarily focuses on converting environments captured with 3d Gaussian Splatting (3DGS) but also supports scenes captured with photogrammetry and LiDAR. It incorporates 3d semantic features, object segmentation, predictive per-gaussian physics properties, object decomposition, deformability, and articulation. To increase policy generalization, it is also able to generate “digital sisters” or similar versions of each object within an environment. We are still evaluating how well policies trained in this environment perform in the real world, and initial results show promising success rates similar to or improving on other recent papers such as SimFoundry and PolaRiS.\",\"author\":[{\"@type\":\"Person\",\"name\":\"Luke Hollis\",\"affiliation\":{\"@type\":\"Organization\",\"name\":\"Harvard Computational Robotics Group, School of Engineering and Applied Sciences, Harvard University\",\"url\":\"https://computationalrobotics.seas.harvard.edu/\"}},{\"@type\":\"Person\",\"name\":\"Tianxing Fan\",\"affiliation\":{\"@type\":\"Organization\",\"name\":\"Harvard Computational Robotics Group, School of Engineering and Applied Sciences, Harvard University\",\"url\":\"https://computationalrobotics.seas.harvard.edu/\"}},{\"@type\":\"Person\",\"name\":\"Heng Yang\",\"affiliation\":{\"@type\":\"Organization\",\"name\":\"Harvard Computational Robotics Group, School of Engineering and Applied Sciences, Harvard University\",\"url\":\"https://computationalrobotics.seas.harvard.edu/\"}}],\"publisher\":{\"@type\":\"Organization\",\"name\":\"Harvard Computational Robotics Group\",\"url\":\"https://computationalrobotics.seas.harvard.edu/\"},\"datePublished\":\"2026\",\"creativeWorkStatus\":\"Preprint\",\"image\":\"https://computationalrobotics.seas.harvard.edu/SceneAgent/branding/sceneagent-social-share.jpg\",\"url\":\"https://computationalrobotics.seas.harvard.edu/SceneAgent/\",\"mainEntityOfPage\":\"https://computationalrobotics.seas.harvard.edu/SceneAgent/\",\"inLanguage\":\"en\",\"keywords\":\"SceneAgent, real-to-sim, sim-to-real, 3D Gaussian Splatting, simulation environments, predictive physics, robot policy evaluation, policy training, online planning, object segmentation, articulation, deformable objects, digital sisters, robot learning, Harvard Computational Robotics Group\"}"])</script><script>self.__next_f.push([1,"0:{\"P\":null,\"c\":[\"\",\"_not-found\",\"\"],\"q\":\"\",\"i\":false,\"f\":[[[\"\",{\"children\":[\"/_not-found\",{\"children\":[\"__PAGE__\",{}]}]},\"$undefined\",\"$undefined\",16],[[\"$\",\"$1\",\"c\",{\"children\":[[[\"$\",\"link\",\"0\",{\"rel\":\"stylesheet\",\"href\":\"./_next/static/chunks/1zf8qs8h9vuqk.css\",\"precedence\":\"next\",\"crossOrigin\":\"$undefined\",\"nonce\":\"$undefined\"}],[\"$\",\"script\",\"script-0\",{\"src\":\"./_next/static/chunks/3d-g45tvvtwds.js\",\"async\":true,\"nonce\":\"$undefined\"}]],[\"$\",\"html\",null,{\"lang\":\"en\",\"className\":\"silkscreen_5c214b00-module__BDnqEG__variable plexmono_301f1454-module__3YJGwG__variable\",\"children\":[\"$\",\"body\",null,{\"children\":[[\"$\",\"$L2\",null,{\"parallelRouterKey\":\"children\",\"error\":\"$undefined\",\"errorStyles\":\"$undefined\",\"errorScripts\":\"$undefined\",\"template\":[\"$\",\"$L3\",null,{}],\"templateStyles\":\"$undefined\",\"templateScripts\":\"$undefined\",\"notFound\":[[[\"$\",\"title\",null,{\"children\":\"404: This page could not be found.\"}],[\"$\",\"div\",null,{\"style\":{\"fontFamily\":\"system-ui,\\\"Segoe UI\\\",Roboto,Helvetica,Arial,sans-serif,\\\"Apple Color Emoji\\\",\\\"Segoe UI Emoji\\\"\",\"height\":\"100vh\",\"textAlign\":\"center\",\"display\":\"flex\",\"flexDirection\":\"column\",\"alignItems\":\"center\",\"justifyContent\":\"center\"},\"children\":[\"$\",\"div\",null,{\"children\":[[\"$\",\"style\",null,{\"dangerouslySetInnerHTML\":{\"__html\":\"body{color:#000;background:#fff;margin:0}.next-error-h1{border-right:1px solid rgba(0,0,0,.3)}@media (prefers-color-scheme:dark){body{color:#fff;background:#000}.next-error-h1{border-right:1px solid rgba(255,255,255,.3)}}\"}}],[\"$\",\"h1\",null,{\"className\":\"next-error-h1\",\"style\":{\"display\":\"inline-block\",\"margin\":\"0 20px 0 0\",\"padding\":\"0 23px 0 0\",\"fontSize\":24,\"fontWeight\":500,\"verticalAlign\":\"top\",\"lineHeight\":\"49px\"},\"children\":404}],[\"$\",\"div\",null,{\"style\":{\"display\":\"inline-block\"},\"children\":[\"$\",\"h2\",null,{\"style\":{\"fontSize\":14,\"fontWeight\":400,\"lineHeight\":\"49px\",\"margin\":0},\"children\":\"This page could not be found.\"}]}]]}]}]],[]],\"forbidden\":\"$undefined\",\"unauthorized\":\"$undefined\"}],[\"$\",\"script\",null,{\"type\":\"application/ld+json\",\"dangerouslySetInnerHTML\":{\"__html\":\"$4\"}}]]}]}]]}],{\"children\":[\"$L5\",{\"children\":[\"$L6\",{},null,false,null]},null,false,\"$@7\"]},null,false,null],\"$L8\",false]],\"m\":\"$undefined\",\"G\":[\"$9\",[\"$La\"]],\"S\":true,\"h\":null,\"s\":\"$undefined\",\"l\":\"$undefined\",\"p\":\"$undefined\",\"d\":\"$undefined\",\"b\":\"F18sWxiUtr6B58sAiA9h4\"}\n"])</script><script>self.__next_f.push([1,"b:I[36215,[\"./_next/static/chunks/3d-g45tvvtwds.js\"],\"OutletBoundary\"]\nc:\"$Sreact.suspense\"\nf:I[36215,[\"./_next/static/chunks/3d-g45tvvtwds.js\"],\"ViewportBoundary\"]\n11:I[36215,[\"./_next/static/chunks/3d-g45tvvtwds.js\"],\"MetadataBoundary\"]\n5:[\"$\",\"$1\",\"c\",{\"children\":[null,[\"$\",\"$L2\",null,{\"parallelRouterKey\":\"children\",\"error\":\"$undefined\",\"errorStyles\":\"$undefined\",\"errorScripts\":\"$undefined\",\"template\":[\"$\",\"$L3\",null,{}],\"templateStyles\":\"$undefined\",\"templateScripts\":\"$undefined\",\"notFound\":\"$undefined\",\"forbidden\":\"$undefined\",\"unauthorized\":\"$undefined\"}]]}]\n6:[\"$\",\"$1\",\"c\",{\"children\":[[[\"$\",\"title\",null,{\"children\":\"404: This page could not be found.\"}],[\"$\",\"div\",null,{\"style\":\"$0:f:0:1:0:props:children:1:props:children:props:children:0:props:notFound:0:1:props:style\",\"children\":[\"$\",\"div\",null,{\"children\":[[\"$\",\"style\",null,{\"dangerouslySetInnerHTML\":{\"__html\":\"body{color:#000;background:#fff;margin:0}.next-error-h1{border-right:1px solid rgba(0,0,0,.3)}@media (prefers-color-scheme:dark){body{color:#fff;background:#000}.next-error-h1{border-right:1px solid rgba(255,255,255,.3)}}\"}}],[\"$\",\"h1\",null,{\"className\":\"next-error-h1\",\"style\":\"$0:f:0:1:0:props:children:1:props:children:props:children:0:props:notFound:0:1:props:children:props:children:1:props:style\",\"children\":404}],[\"$\",\"div\",null,{\"style\":\"$0:f:0:1:0:props:children:1:props:children:props:children:0:props:notFound:0:1:props:children:props:children:2:props:style\",\"children\":[\"$\",\"h2\",null,{\"style\":\"$0:f:0:1:0:props:children:1:props:children:props:children:0:props:notFound:0:1:props:children:props:children:2:props:children:props:style\",\"children\":\"This page could not be found.\"}]}]]}]}]],null,[\"$\",\"$Lb\",null,{\"children\":[\"$\",\"$c\",null,{\"name\":\"Next.MetadataOutlet\",\"children\":\"$@d\"}]}]]}]\ne:[]\n7:\"$We\"\n8:[\"$\",\"$1\",\"h\",{\"children\":[[\"$\",\"meta\",null,{\"name\":\"robots\",\"content\":\"noindex\"}],[\"$\",\"$Lf\",null,{\"children\":\"$L10\"}],[\"$\",\"div\",null,{\"hidden\":true,\"children\":[\"$\",\"$L11\",null,{\"children\":[\"$\",\"$c\",null,{\"name\":\"Next.Metadata\",\"children\":\"$L12\"}]}]}],[\"$\",\"meta\",null,{\"name\":\"next-size-adjust\",\"content\":\"\"}]]}]\na:[\"$\",\"link\",\"0\",{\"rel\":\"stylesheet\",\"href\":\"./_next/static/chunks/1zf8qs8h9vuqk.css\",\"precedence\":\"next\",\"crossOrigin\":\"$undefined\",\"nonce\":\"$undefined\"}]\n"])</script><script>self.__next_f.push([1,"10:[[\"$\",\"meta\",\"0\",{\"charSet\":\"utf-8\"}],[\"$\",\"meta\",\"1\",{\"name\":\"viewport\",\"content\":\"width=device-width, initial-scale=1\"}]]\n"])</script><script>self.__next_f.push([1,"d:null\n"])</script><script>self.__next_f.push([1,"12:[[\"$\",\"title\",\"0\",{\"children\":\"SceneAgent | Harvard Computational Robotics Group\"}],[\"$\",\"meta\",\"1\",{\"name\":\"description\",\"content\":\"SceneAgent converts 3D captures of real-world scenes into simulation environments with predictive physics for robot policy evaluation, planning and training.\"}],[\"$\",\"link\",\"2\",{\"rel\":\"author\",\"href\":\"https://computationalrobotics.seas.harvard.edu/\"}],[\"$\",\"meta\",\"3\",{\"name\":\"author\",\"content\":\"Luke Hollis\"}],[\"$\",\"link\",\"4\",{\"rel\":\"author\",\"href\":\"https://computationalrobotics.seas.harvard.edu/\"}],[\"$\",\"meta\",\"5\",{\"name\":\"author\",\"content\":\"Tianxing Fan\"}],[\"$\",\"link\",\"6\",{\"rel\":\"author\",\"href\":\"https://computationalrobotics.seas.harvard.edu/\"}],[\"$\",\"meta\",\"7\",{\"name\":\"author\",\"content\":\"Heng Yang\"}],[\"$\",\"meta\",\"8\",{\"name\":\"keywords\",\"content\":\"SceneAgent,real-to-sim,sim-to-real,3D Gaussian Splatting,simulation environments,predictive physics,robot policy evaluation,policy training,online planning,object segmentation,articulation,deformable objects,digital sisters,robot learning,Harvard Computational Robotics Group\"}],[\"$\",\"meta\",\"9\",{\"name\":\"creator\",\"content\":\"Harvard Computational Robotics Group\"}],[\"$\",\"meta\",\"10\",{\"name\":\"publisher\",\"content\":\"Harvard Computational Robotics Group\"}],[\"$\",\"meta\",\"11\",{\"name\":\"robots\",\"content\":\"index, follow\"}],[\"$\",\"meta\",\"12\",{\"name\":\"googlebot\",\"content\":\"index, follow, max-video-preview:-1, max-image-preview:large, max-snippet:-1\"}],[\"$\",\"meta\",\"13\",{\"name\":\"citation_title\",\"content\":\"SceneAgent: 3D Capture-Derived Scenes with Predictive Physics for Policy Evaluation and Training Environments\"}],[\"$\",\"meta\",\"14\",{\"name\":\"citation_author\",\"content\":\"Luke Hollis\"}],[\"$\",\"meta\",\"15\",{\"name\":\"citation_author\",\"content\":\"Tianxing Fan\"}],[\"$\",\"meta\",\"16\",{\"name\":\"citation_author\",\"content\":\"Heng Yang\"}],[\"$\",\"meta\",\"17\",{\"name\":\"citation_publication_date\",\"content\":\"2026\"}],[\"$\",\"meta\",\"18\",{\"name\":\"citation_publisher\",\"content\":\"Harvard Computational Robotics Group\"}],[\"$\",\"meta\",\"19\",{\"name\":\"citation_abstract_html_url\",\"content\":\"https://computationalrobotics.seas.harvard.edu/SceneAgent/\"}],[\"$\",\"link\",\"20\",{\"rel\":\"canonical\",\"href\":\"https://computationalrobotics.seas.harvard.edu/SceneAgent/\"}],[\"$\",\"meta\",\"21\",{\"property\":\"og:title\",\"content\":\"SceneAgent: 3D captures to simulation environments with predictive physics\"}],[\"$\",\"meta\",\"22\",{\"property\":\"og:description\",\"content\":\"An agentic pipeline from 3D Gaussian Splatting, photogrammetry and LiDAR captures of real scenes to simulation environments with predictive physics. Harvard Computational Robotics Group, 2026 preprint.\"}],[\"$\",\"meta\",\"23\",{\"property\":\"og:url\",\"content\":\"https://computationalrobotics.seas.harvard.edu/SceneAgent/\"}],[\"$\",\"meta\",\"24\",{\"property\":\"og:site_name\",\"content\":\"Harvard Computational Robotics Group\"}],[\"$\",\"meta\",\"25\",{\"property\":\"og:locale\",\"content\":\"en_US\"}],[\"$\",\"meta\",\"26\",{\"property\":\"og:image\",\"content\":\"https://computationalrobotics.seas.harvard.edu/SceneAgent/branding/sceneagent-social-share.jpg\"}],[\"$\",\"meta\",\"27\",{\"property\":\"og:image:secure_url\",\"content\":\"https://computationalrobotics.seas.harvard.edu/SceneAgent/branding/sceneagent-social-share.jpg\"}],[\"$\",\"meta\",\"28\",{\"property\":\"og:image:type\",\"content\":\"image/jpeg\"}],[\"$\",\"meta\",\"29\",{\"property\":\"og:image:width\",\"content\":\"1200\"}],[\"$\",\"meta\",\"30\",{\"property\":\"og:image:height\",\"content\":\"630\"}],[\"$\",\"meta\",\"31\",{\"property\":\"og:image:alt\",\"content\":\"SceneAgent: two Franka robot arms over a simulated tabletop scene, above the paper title and subtitle\"}],[\"$\",\"meta\",\"32\",{\"property\":\"og:type\",\"content\":\"website\"}],[\"$\",\"meta\",\"33\",{\"name\":\"twitter:card\",\"content\":\"summary_large_image\"}],[\"$\",\"meta\",\"34\",{\"name\":\"twitter:title\",\"content\":\"SceneAgent: 3D captures to simulation environments with predictive physics\"}],[\"$\",\"meta\",\"35\",{\"name\":\"twitter:description\",\"content\":\"An agentic pipeline from 3D Gaussian Splatting, photogrammetry and LiDAR captures of real scenes to simulation environments with predictive physics. Harvard Computational Robotics Group, 2026 preprint.\"}],\"$L13\",\"$L14\",\"$L15\",\"$L16\"]\n"])</script><script>self.__next_f.push([1,"17:I[29447,[\"./_next/static/chunks/3d-g45tvvtwds.js\"],\"IconMark\"]\n13:[\"$\",\"meta\",\"36\",{\"name\":\"twitter:image\",\"content\":\"https://computationalrobotics.seas.harvard.edu/SceneAgent/branding/sceneagent-social-share.jpg\"}]\n14:[\"$\",\"meta\",\"37\",{\"name\":\"twitter:image:alt\",\"content\":\"SceneAgent: two Franka robot arms over a simulated tabletop scene, above the paper title and subtitle\"}]\n15:[\"$\",\"link\",\"38\",{\"rel\":\"icon\",\"href\":\"./favicon.png\",\"type\":\"image/png\"}]\n16:[\"$\",\"$L17\",\"39\",{}]\n"])</script></body></html>