[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"branding":3,"analytics":7,"article-teaching-ai-to-see-in-3d-using-only-flat-video":10,"sections":34},{"siteName":4,"siteTagline":5,"publisherName":4,"contactEmail":6},"The Revision","Tech news, decoded.","editor@therevision.news",{"gaMeasurementId":8,"adsenseClientId":9},"G-ZW2MV82GYR","ca-pub-8533917693782264",{"article":11},{"id":12,"slug":13,"title":14,"dek":15,"body_md":16,"tags_json":17,"published_at":18,"created_at":19,"updated_at":20,"status":21,"review_note":22,"review_notes":23,"image_url":22,"persona_id":22,"persona_name":22,"section":24,"tags":25,"sources":29,"feedback":33,"feedback_at":22,"cost_usd":33,"total_tokens":33},1811,"teaching-ai-to-see-in-3d-using-only-flat-video","Teaching AI to See in 3D Using Only Flat Video","A new training framework called GeoVR gives multimodal AI models spatial awareness by extracting geometry from ordinary 2D video, no 3D data required.","Multimodal AI models can describe a photo in detail but struggle to reason about depth, distance, or how objects relate in physical space.\n\nResearchers introduced GeoVR, a framework that trains large language models to develop 3D spatial awareness using nothing but ordinary 2D video sequences. Rather than requiring scarce labeled 3D datasets, GeoVR distills geometry knowledge from pre-trained 3D models and uses it to reshape how the AI internally represents what it sees. The system pursues four simultaneous learning targets: estimating how the camera moved between frames, building dense depth maps, calculating a real-world scale factor, and aligning intermediate features with 3D space. Tested on spatial reasoning benchmarks, the approach reached state-of-the-art results.\n\nThe core insight matters because 3D training data is genuinely hard to collect at scale, while video is everywhere. If a model can learn geometric intuition from YouTube-style footage, the barrier to building spatially aware AI drops considerably. That has obvious implications for robotics, AR, and any application where understanding physical layout is non-negotiable.\n\nThe catch is that \"state-of-the-art on benchmarks\" and \"actually understands space like a person does\" are not the same sentence — a distinction the field has had to relearn more than once.","[\"ai\",\"computer vision\",\"multimodal\",\"spatial reasoning\"]","2026-06-19T04:00:00.000Z","2026-06-19T12:26:02.845Z","2026-06-19T14:22:19.854Z","published",null,[],"ai",[24,26,27,28],"computer vision","multimodal","spatial reasoning",[30],{"name":31,"url":32},"arXiv cs.AI","https:\u002F\u002Farxiv.org\u002Fabs\u002F2606.05833",0,{"sections":35},[36,40,44,49,54,59,64,68,72,77,82,87,92,97],{"name":37,"slug":24,"count":38,"latest_published_at":39},"AI",491,"2026-06-19T14:59:11.000Z",{"name":41,"slug":42,"count":43,"latest_published_at":18},"Security","security",132,{"name":45,"slug":46,"count":47,"latest_published_at":48},"Policy","policy",88,"2026-06-16T09:26:09.000Z",{"name":50,"slug":51,"count":52,"latest_published_at":53},"Consumer Tech","consumer-tech",78,"2026-06-16T17:58:24.000Z",{"name":55,"slug":56,"count":57,"latest_published_at":58},"Hardware","hardware",62,"2026-06-18T15:24:16.000Z",{"name":60,"slug":61,"count":62,"latest_published_at":63},"Deals","deals",58,"2026-06-19T14:43:50.000Z",{"name":65,"slug":66,"count":62,"latest_published_at":67},"Software","software","2026-06-16T20:00:00.000Z",{"name":69,"slug":70,"count":71,"latest_published_at":18},"Dev Tools","dev-tools",50,{"name":73,"slug":74,"count":75,"latest_published_at":76},"Science","science",38,"2026-06-18T04:00:00.000Z",{"name":78,"slug":79,"count":80,"latest_published_at":81},"Gaming","gaming",31,"2026-06-16T15:25:13.000Z",{"name":83,"slug":84,"count":85,"latest_published_at":86},"General","general",26,"2026-06-13T18:35:15.000Z",{"name":88,"slug":89,"count":90,"latest_published_at":91},"Startups","startups",23,"2026-06-16T15:00:00.000Z",{"name":93,"slug":94,"count":95,"latest_published_at":96},"Reviews","reviews",19,"2026-06-14T08:00:00.000Z",{"name":98,"slug":99,"count":100,"latest_published_at":101},"How-To","how-to",6,"2026-06-16T09:00:00.000Z"]