<video src=\"https://cdn-uploads.huggingface.co/production/uploads/64796780bb9a5693c48c97c6/As_R58YJBKFaLEEl_Lyu2.mp4\" controls=\"\" class=\"max-w-full!\"></video></p>","updatedAt":"2026-05-21T03:10:52.008Z","author":{"_id":"64796780bb9a5693c48c97c6","avatarUrl":"/avatars/1226b17dc24265d863fb99befe0d2187.svg","fullname":"jjr5401","name":"JiaJinrang","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.5726121664047241},"editors":["JiaJinrang"],"editorAvatarUrls":["/avatars/1226b17dc24265d863fb99befe0d2187.svg"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2605.17916","authors":[{"_id":"6a0d1a5265eb30f20d962c00","user":{"_id":"64796780bb9a5693c48c97c6","avatarUrl":"/avatars/1226b17dc24265d863fb99befe0d2187.svg","isPro":false,"fullname":"jjr5401","user":"JiaJinrang","type":"user","name":"JiaJinrang"},"name":"Jinrang Jia","status":"claimed_verified","statusLastChangedAt":"2026-05-20T17:11:24.716Z","hidden":false},{"_id":"6a0d1a5265eb30f20d962c01","name":"Zhenjia Li","hidden":false},{"_id":"6a0d1a5265eb30f20d962c02","name":"Yijiang Hu","hidden":false},{"_id":"6a0d1a5265eb30f20d962c03","name":"Yifeng Shi","hidden":false}],"publishedAt":"2026-05-19T00:00:00.000Z","submittedOnDailyAt":"2026-05-21T00:00:00.000Z","title":"PanoWorld: A Generative Spatial World Model for Consistent Whole-House Panorama Synthesis","submittedOnDailyBy":{"_id":"64796780bb9a5693c48c97c6","avatarUrl":"/avatars/1226b17dc24265d863fb99befe0d2187.svg","isPro":false,"fullname":"jjr5401","user":"JiaJinrang","type":"user","name":"JiaJinrang"},"summary":"Generating a consistent whole-house VR tour from a floorplan and style reference requires both photorealistic panoramas and cross-view spatial coherence. Pure 2D generators produce appealing single panoramas but re-imagine geometry and materials when the viewpoint changes, whereas monolithic 3D generation becomes expensive and loses fine texture at multi-room scale. We introduce PanoWorld, a generative spatial world model that treats whole-house synthesis as autoregressive generation of node-based 360-degree panoramas, matching the discrete navigation used by real VR tour products. PanoWorld uses a floorplan-derived 3D shell as a global geometric proxy and a dynamic 3D Gaussian Splatting cache as renderable spatial memory. A feed-forward panoramic LRM designed for metric-scale multi-room 360-degree inputs lifts generated panoramas into local 3DGS updates, while Room-aware Group Attention suppresses cross-room feature interference. A topology-aware progressive caching strategy fuses these local updates without repeatedly reconstructing the full history. By decoupling shell-based geometry guidance from cache-rendered visual memory, PanoWorld preserves high-frequency 2D synthesis quality while improving cross-node layout and material consistency. The project link is https://jjrcn.github.io/PanoWorld-project-home/","upvotes":4,"discussionId":"6a0d1a5265eb30f20d962c04","ai_summary":"PanoWorld generates consistent VR tours by combining 3D geometric guidance with dynamic visual memory, enabling high-quality multi-room panoramas with spatial coherence.","ai_keywords":["panoramic LRM","3D Gaussian Splatting","Room-aware Group Attention","topology-aware progressive caching","autoregressive generation","360-degree panoramas","geometric proxy","spatial memory"]},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"64796780bb9a5693c48c97c6","avatarUrl":"/avatars/1226b17dc24265d863fb99befe0d2187.svg","isPro":false,"fullname":"jjr5401","user":"JiaJinrang","type":"user"},{"_id":"67061c5304a5c834862dae98","avatarUrl":"/avatars/e0a63ac71aa6fe7fe96bbcc250ab0f25.svg","isPro":false,"fullname":"Shi","user":"YiFeng2933","type":"user"},{"_id":"64b8de642fccad9f5f0a4ce4","avatarUrl":"/avatars/be62765ece88b0a6517753d1329f2acb.svg","isPro":false,"fullname":"ismail codar","user":"ismail-codar","type":"user"},{"_id":"6407e5294edf9f5c4fd32228","avatarUrl":"/avatars/8e2d55460e9fe9c426eb552baf4b2cb0.svg","isPro":false,"fullname":"Stoney Kang","user":"sikang99","type":"user"}],"acceptLanguages":["en"],"dailyPaperRank":0,"markdownContentUrl":"https://huggingface.co/buckets/huggingchat/papers-content/resolve/2605/2605.17916.md"}">

Get this paper in your agent:

Don't have the latest CLI?

Cite arxiv.org/abs/2605.17916 in a model README.md to link it from this page.

Cite arxiv.org/abs/2605.17916 in a dataset README.md to link it from this page.

Cite arxiv.org/abs/2605.17916 in a Space README.md to link it from this page.

Add this paper to a collection to link it from this page.