Recovers part geometry and joint parameters from a single resting-state observation, exported directly as a simulation-ready URDF.</p>\n","updatedAt":"2026-08-12T06:00:20.850Z","author":{"_id":"67e4ae9f765e3105ef0928c3","avatarUrl":"/avatars/db5401580376f081dab866c7bf383207.svg","fullname":"Daeun Lee","name":"da-eun07","type":"user","isPro":true,"isHf":false,"isHfAdmin":false,"isMod":false,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.8182973265647888},"editors":["da-eun07"],"editorAvatarUrls":["/avatars/db5401580376f081dab866c7bf383207.svg"],"reactions":[],"isReport":false}},{"id":"6a7c43422b64685499153db5","author":{"_id":"6a7c419477297c7f9f56dc2e","avatarUrl":"/avatars/619f76b34c533dd9fc73d217bae007f6.svg","fullname":"sanjyot keer","name":"sanjyot011keer","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"isUserFollowing":false},"createdAt":"2026-08-12T09:56:18.000Z","type":"comment","data":{"edited":false,"hidden":false,"latest":{"raw":"a fascinating approach to articulated 3D reconstruction. The idea of using a single rest-state image and then leveraging video diffusion to generate articulation hypotheses is especially interesting. ","html":"<p>a fascinating approach to articulated 3D reconstruction. The idea of using a single rest-state image and then leveraging video diffusion to generate articulation hypotheses is especially interesting. </p>\n","updatedAt":"2026-08-12T09:56:18.526Z","author":{"_id":"6a7c419477297c7f9f56dc2e","avatarUrl":"/avatars/619f76b34c533dd9fc73d217bae007f6.svg","fullname":"sanjyot keer","name":"sanjyot011keer","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.8406922221183777},"editors":["sanjyot011keer"],"editorAvatarUrls":["/avatars/619f76b34c533dd9fc73d217bae007f6.svg"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2607.27749","authors":[{"_id":"6a7c08c71653ef87c6af1d83","name":"Daeun Lee","hidden":false},{"_id":"6a7c08c71653ef87c6af1d84","name":"Jaeah Lee","hidden":false},{"_id":"6a7c08c71653ef87c6af1d85","name":"Woosung Kim","hidden":false},{"_id":"6a7c08c71653ef87c6af1d86","name":"Haebeom Jung","hidden":false},{"_id":"6a7c08c71653ef87c6af1d87","name":"Jaesik Park","hidden":false}],"mediaUrls":["https://cdn-uploads.huggingface.co/production/uploads/67e4ae9f765e3105ef0928c3/XoXKw-kio3zyIWYwxix1o.mp4"],"publishedAt":"2026-07-30T00:00:00.000Z","submittedOnDailyAt":"2026-08-12T00:00:00.000Z","title":"Articulated Object Reconstruction from Rest-State Observation","submittedOnDailyBy":{"_id":"67e4ae9f765e3105ef0928c3","avatarUrl":"/avatars/db5401580376f081dab866c7bf383207.svg","isPro":true,"fullname":"Daeun Lee","user":"da-eun07","type":"user","name":"da-eun07"},"summary":"Building interactive digital twins requires recovering both 3D geometry and the kinematic structures that govern how objects articulate. Yet existing methods for articulated object reconstruction require explicitly observable motion from multiple articulation states. We introduce a rest-state formulation that reconstructs articulated objects from a single closed configuration, an inherently ill-posed setting where geometry, semantics, and motion priors compensate for the absence of motion cues. Our framework adopts an explicit mesh as an intermediate representation for cross-model verification and fusion, reconciling noisy outputs from vision-language and segmentation models into spatially consistent part structures. To estimate joint parameters without observed motion, we use a video diffusion model to synthesize articulation hypotheses and validate them through geometric consistency. Our approach achieves accurate part decomposition and physically plausible articulation, performing competitively with motion-observing reconstruction-based, generation-based, and modular pretrained-model baselines.","upvotes":33,"discussionId":"6a7c08c71653ef87c6af1d88","projectPage":"https://da-eun07.github.io/rest2art/","ai_summary":"A rest-state framework reconstructs articulated objects from a single closed configuration by fusing vision-language outputs into consistent part meshes and validating synthesized motion hypotheses via geometric consistency.","ai_keywords":["digital twins","articulated object reconstruction","rest-state formulation","explicit mesh","vision-language models","segmentation models","video diffusion model","articulation hypotheses","geometric consistency","joint parameters"],"ai_summary_model":"thinkingmachines/Inkling-Small","organization":{"_id":"6304ac9f5d136debcecba4fd","name":"snu","fullname":"Seoul National University","avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/6304abe86dbbb80f1636b20b/5zYAx1sdCvxw3EV1zaIbf.png"}},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"66af12ced7f2b182861c5ffe","avatarUrl":"/avatars/ca81daa9ae6ae437045e56968e7953a9.svg","isPro":false,"fullname":"Hyunjoon Lee","user":"jjoonii","type":"user"},{"_id":"66670c59a1f25968c76486df","avatarUrl":"/avatars/394217c86ab926274cf80c5c49f7bb22.svg","isPro":false,"fullname":"Haebeom Jung","user":"b1o1o1m","type":"user"},{"_id":"636f9ff831af06da864ba186","avatarUrl":"/avatars/0a9149746ac5279158ae21934fa06f8b.svg","isPro":false,"fullname":"cubec","user":"cubec","type":"user"},{"_id":"66728afc4de1918e09192525","avatarUrl":"/avatars/9a2d444f0065cbdb5e40420e2a7cc64b.svg","isPro":true,"fullname":"Seungwoo Yoon","user":"SW-Yoon","type":"user"},{"_id":"66270c20846c2a66a7a7c299","avatarUrl":"/avatars/0cddccc9584b1062faaff164f0de6e22.svg","isPro":true,"fullname":"Sangmin","user":"nstar1125","type":"user"},{"_id":"6863859733f038ee25d9d1e2","avatarUrl":"/avatars/068b2e38dec86346869f6120a5cad748.svg","isPro":false,"fullname":"Kyoosuk, HWANG","user":"kyoosuk","type":"user"},{"_id":"67298cbd9c7db29615b1b5b9","avatarUrl":"/avatars/ae5b7e7bb27bf68c2916c7f77c9bb359.svg","isPro":false,"fullname":"Hoyong Kim","user":"khyong","type":"user"},{"_id":"67e4ae9f765e3105ef0928c3","avatarUrl":"/avatars/db5401580376f081dab866c7bf383207.svg","isPro":true,"fullname":"Daeun Lee","user":"da-eun07","type":"user"},{"_id":"68177af0a6da0279eb9a43b3","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/noauth/nBbypIUh2e5LLiTz7kuNb.png","isPro":false,"fullname":"Soonho Kim","user":"suNoeul","type":"user"},{"_id":"69d326e39b543e9d25c99c2b","avatarUrl":"/avatars/bb8cb301001794dc88991a6352f4f459.svg","isPro":false,"fullname":"sdfwesd","user":"wadfwef","type":"user"},{"_id":"69cc9f75a3e592deb964a118","avatarUrl":"/avatars/0a7b3e3275ef8c7d270d9ccc2691c17a.svg","isPro":false,"fullname":"Sunghyun Choi","user":"Sunghyun12333","type":"user"},{"_id":"64a75d8374cc37096692593d","avatarUrl":"/avatars/295c1bbf9c3e8a13fbd3337c3ac0e86f.svg","isPro":false,"fullname":"Pierre Musacchio","user":"pmusacchio","type":"user"}],"acceptLanguages":["en"],"dailyPaperRank":0,"organization":{"_id":"6304ac9f5d136debcecba4fd","name":"snu","fullname":"Seoul National University","avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/6304abe86dbbb80f1636b20b/5zYAx1sdCvxw3EV1zaIbf.png"},"query":{}}">

Cite arxiv.org/abs/2607.27749 in a model README.md to link it from this page.

Cite arxiv.org/abs/2607.27749 in a dataset README.md to link it from this page.

Cite arxiv.org/abs/2607.27749 in a Space README.md to link it from this page.

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