https://x.com/Luxi_Chen123/status/1902280142812242151
Arxiv Paper：https://arxiv.org/abs/2503.13265
Project：https://ml-gsai.github.io/FlexWorld/
Github：https://github.com/ML-GSAI/FlexWorld

All Code and Weights are Open-sourced. Welcome to enjoy FlexWorld!

\n","updatedAt":"2025-03-19T10:39:30.823Z","author":{"_id":"638efcf4c67af472d316d424","avatarUrl":"/avatars/97a57859d7d87a3a8f1bb41d32a72bc2.svg","fullname":"Ge Zhang","name":"zhangysk","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":77,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.39469674229621887},"editors":["zhangysk"],"editorAvatarUrls":["/avatars/97a57859d7d87a3a8f1bb41d32a72bc2.svg"],"reactions":[{"reaction":"🔥","users":["AdinaY"],"count":1},{"reaction":"🚀","users":["AdinaY"],"count":1}],"isReport":false}},{"id":"67db70b02aeb1103caf092cf","author":{"_id":"63d3e0e8ff1384ce6c5dd17d","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1674830754237-63d3e0e8ff1384ce6c5dd17d.jpeg","fullname":"Librarian Bot (Bot)","name":"librarian-bot","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":318,"isUserFollowing":false},"createdAt":"2025-03-20T01:34:40.000Z","type":"comment","data":{"edited":false,"hidden":false,"latest":{"raw":"This is an automated message from the [Librarian Bot](https://huggingface.co/librarian-bots). I found the following papers similar to this paper. \n\nThe following papers were recommended by the Semantic Scholar API \n\n* [InsTex: Indoor Scenes Stylized Texture Synthesis](https://huggingface.co/papers/2501.13969) (2025)\n* [AuraFusion360: Augmented Unseen Region Alignment for Reference-based 360° Unbounded Scene Inpainting](https://huggingface.co/papers/2502.05176) (2025)\n* [V2Edit: Versatile Video Diffusion Editor for Videos and 3D Scenes](https://huggingface.co/papers/2503.10634) (2025)\n* [Enhancing Monocular 3D Scene Completion with Diffusion Model](https://huggingface.co/papers/2503.00726) (2025)\n* [Generative Gaussian Splatting: Generating 3D Scenes with Video Diffusion Priors](https://huggingface.co/papers/2503.13272) (2025)\n* [WonderVerse: Extendable 3D Scene Generation with Video Generative Models](https://huggingface.co/papers/2503.09160) (2025)\n* [CineMaster: A 3D-Aware and Controllable Framework for Cinematic Text-to-Video Generation](https://huggingface.co/papers/2502.08639) (2025)\n\n\n Please give a thumbs up to this comment if you found it helpful!\n\n If you want recommendations for any Paper on Hugging Face checkout [this](https://huggingface.co/spaces/librarian-bots/recommend_similar_papers) Space\n\n You can directly ask Librarian Bot for paper recommendations by tagging it in a comment: `@librarian-bot recommend`","html":"

This is an automated message from the Librarian Bot. I found the following papers similar to this paper.

The following papers were recommended by the Semantic Scholar API

Please give a thumbs up to this comment if you found it helpful!

If you want recommendations for any Paper on Hugging Face checkout this Space

You can directly ask Librarian Bot for paper recommendations by tagging it in a comment: \n\n@librarian-bot\n\t recommend

\n","updatedAt":"2025-03-20T01:34:40.894Z","author":{"_id":"63d3e0e8ff1384ce6c5dd17d","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1674830754237-63d3e0e8ff1384ce6c5dd17d.jpeg","fullname":"Librarian Bot (Bot)","name":"librarian-bot","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":318,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.7007559537887573},"editors":["librarian-bot"],"editorAvatarUrls":["https://cdn-avatars.huggingface.co/v1/production/uploads/1674830754237-63d3e0e8ff1384ce6c5dd17d.jpeg"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2503.13265","authors":[{"_id":"67da8c561a87aaeb4ca06067","user":{"_id":"67a2c1eb07690f2a570edb15","avatarUrl":"/avatars/b4a56782774a590ef01b1a1ec96464b2.svg","isPro":false,"fullname":"luxi chen","user":"lucydetrack","type":"user"},"name":"Luxi Chen","status":"admin_assigned","statusLastChangedAt":"2025-03-19T13:39:20.898Z","hidden":false},{"_id":"67da8c561a87aaeb4ca06068","user":{"_id":"64586db4bb7bdc9655afb5ce","avatarUrl":"/avatars/0756344c48927a3b92e583fe6a43868c.svg","isPro":false,"fullname":"Zhou","user":"ZihanZhou","type":"user"},"name":"Zihan Zhou","status":"admin_assigned","statusLastChangedAt":"2025-03-19T13:39:28.536Z","hidden":false},{"_id":"67da8c561a87aaeb4ca06069","name":"Min Zhao","hidden":false},{"_id":"67da8c561a87aaeb4ca0606a","name":"Yikai Wang","hidden":false},{"_id":"67da8c561a87aaeb4ca0606b","user":{"_id":"638efcf4c67af472d316d424","avatarUrl":"/avatars/97a57859d7d87a3a8f1bb41d32a72bc2.svg","isPro":false,"fullname":"Ge Zhang","user":"zhangysk","type":"user"},"name":"Ge Zhang","status":"admin_assigned","statusLastChangedAt":"2025-03-19T13:40:15.137Z","hidden":false},{"_id":"67da8c561a87aaeb4ca0606c","name":"Wenhao Huang","hidden":false},{"_id":"67da8c561a87aaeb4ca0606d","name":"Hao Sun","hidden":false},{"_id":"67da8c561a87aaeb4ca0606e","user":{"_id":"64b8c89052b7353d8c6a1013","avatarUrl":"/avatars/cd59fffe81f6b07b4519540b8ff3d95f.svg","isPro":false,"fullname":"Ji-Rong Wen","user":"jrwen","type":"user"},"name":"Ji-Rong Wen","status":"admin_assigned","statusLastChangedAt":"2025-03-19T13:39:41.490Z","hidden":false},{"_id":"67da8c561a87aaeb4ca0606f","user":{"_id":"64c07b488e2612254361153b","avatarUrl":"/avatars/ade0f783cc4c2d3e73f402637f595471.svg","isPro":false,"fullname":"chongxuan li","user":"zhenxuan00","type":"user"},"name":"Chongxuan Li","status":"admin_assigned","statusLastChangedAt":"2025-03-19T13:39:35.810Z","hidden":false}],"publishedAt":"2025-03-17T15:18:38.000Z","submittedOnDailyAt":"2025-03-19T09:09:30.790Z","title":"FlexWorld: Progressively Expanding 3D Scenes for Flexiable-View\n Synthesis","submittedOnDailyBy":{"_id":"638efcf4c67af472d316d424","avatarUrl":"/avatars/97a57859d7d87a3a8f1bb41d32a72bc2.svg","isPro":false,"fullname":"Ge Zhang","user":"zhangysk","type":"user"},"summary":"Generating flexible-view 3D scenes, including 360{\\deg} rotation and zooming,\nfrom single images is challenging due to a lack of 3D data. To this end, we\nintroduce FlexWorld, a novel framework consisting of two key components: (1) a\nstrong video-to-video (V2V) diffusion model to generate high-quality novel view\nimages from incomplete input rendered from a coarse scene, and (2) a\nprogressive expansion process to construct a complete 3D scene. In particular,\nleveraging an advanced pre-trained video model and accurate depth-estimated\ntraining pairs, our V2V model can generate novel views under large camera pose\nvariations. Building upon it, FlexWorld progressively generates new 3D content\nand integrates it into the global scene through geometry-aware scene fusion.\nExtensive experiments demonstrate the effectiveness of FlexWorld in generating\nhigh-quality novel view videos and flexible-view 3D scenes from single images,\nachieving superior visual quality under multiple popular metrics and datasets\ncompared to existing state-of-the-art methods. Qualitatively, we highlight that\nFlexWorld can generate high-fidelity scenes with flexible views like 360{\\deg}\nrotations and zooming. Project page: https://ml-gsai.github.io/FlexWorld.","upvotes":15,"discussionId":"67da8c601a87aaeb4ca06337","projectPage":"https://ml-gsai.github.io/FlexWorld/","githubRepo":"https://github.com/ML-GSAI/FlexWorld","githubRepoAddedBy":"user","ai_summary":"FlexWorld generates high-quality flexible-view 3D scenes from single images using a video-to-video diffusion model and progressive expansion.","ai_keywords":["video-to-video diffusion model","progressive expansion process","pre-trained video model","depth-estimated training pairs","novel view images","camera pose variations","geometry-aware scene fusion"],"githubStars":129},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"638efcf4c67af472d316d424","avatarUrl":"/avatars/97a57859d7d87a3a8f1bb41d32a72bc2.svg","isPro":false,"fullname":"Ge Zhang","user":"zhangysk","type":"user"},{"_id":"65a374a59acab1998092a9bc","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/65a374a59acab1998092a9bc/M3s_7bSf9G-6b9nLg7N3Z.jpeg","isPro":false,"fullname":"Antonio","user":"JuntingZhou","type":"user"},{"_id":"65bb11cb00a03997849e9e85","avatarUrl":"/avatars/17022b0254192a837f4fe00d84389cda.svg","isPro":false,"fullname":"Wenhao Huang","user":"StephenHuang","type":"user"},{"_id":"64ba096e760936217a3ad2e2","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/64ba096e760936217a3ad2e2/aNQK83Jg5PsBkY0UDg-RA.jpeg","isPro":false,"fullname":"Linzheng Chai","user":"Challenging666","type":"user"},{"_id":"648eb1eb59c4e5c87dc116e0","avatarUrl":"/avatars/c636cea39c2c0937f01398c94ead5dad.svg","isPro":false,"fullname":"fdsqefsgergd","user":"T-representer","type":"user"},{"_id":"63a9b41cc847db253f3f587b","avatarUrl":"/avatars/f9710d6828313d4cfe3819a43e663873.svg","isPro":false,"fullname":"ddddd","user":"yongzhong","type":"user"},{"_id":"665b133508d536a8ac804f7d","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/noauth/Uwi0OnANdTbRbHHQvGqvR.png","isPro":false,"fullname":"Paulson","user":"Pnaomi","type":"user"},{"_id":"64c07b488e2612254361153b","avatarUrl":"/avatars/ade0f783cc4c2d3e73f402637f595471.svg","isPro":false,"fullname":"chongxuan li","user":"zhenxuan00","type":"user"},{"_id":"67db792483cf8143dc320868","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/no-auth/NKGZDKs4LSVGjgYB8WyCz.png","isPro":false,"fullname":"tianshi","user":"anglemg","type":"user"},{"_id":"620783f24e28382272337ba4","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/620783f24e28382272337ba4/zkUveQPNiDfYjgGhuFErj.jpeg","isPro":false,"fullname":"GuoLiangTang","user":"Tommy930","type":"user"},{"_id":"67db865bbc106b53b05e7881","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/no-auth/nW_URK32AWSllGilmZWgC.png","isPro":false,"fullname":"Zixia Wang","user":"SophiaWang","type":"user"},{"_id":"63a369d98c0c89dcae3b8329","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/63a369d98c0c89dcae3b8329/AiH2zjy1cnt9OADAAZMLD.jpeg","isPro":false,"fullname":"Adina Yakefu","user":"AdinaY","type":"user"}],"acceptLanguages":["*"],"dailyPaperRank":0}">

Papers

arxiv:2503.13265

FlexWorld: Progressively Expanding 3D Scenes for Flexiable-View Synthesis

Published on Mar 17, 2025

· Submitted by

Ge Zhang on Mar 19, 2025

Upvote

Authors:

Luxi Chen ,

Zihan Zhou ,

Ge Zhang ,

Ji-Rong Wen ,

Chongxuan Li

Abstract

FlexWorld generates high-quality flexible-view 3D scenes from single images using a video-to-video diffusion model and progressive expansion.

AI-generated summary

Generating flexible-view 3D scenes, including 360{\deg} rotation and zooming, from single images is challenging due to a lack of 3D data. To this end, we introduce FlexWorld, a novel framework consisting of two key components: (1) a strong video-to-video (V2V) diffusion model to generate high-quality novel view images from incomplete input rendered from a coarse scene, and (2) a progressive expansion process to construct a complete 3D scene. In particular, leveraging an advanced pre-trained video model and accurate depth-estimated training pairs, our V2V model can generate novel views under large camera pose variations. Building upon it, FlexWorld progressively generates new 3D content and integrates it into the global scene through geometry-aware scene fusion. Extensive experiments demonstrate the effectiveness of FlexWorld in generating high-quality novel view videos and flexible-view 3D scenes from single images, achieving superior visual quality under multiple popular metrics and datasets compared to existing state-of-the-art methods. Qualitatively, we highlight that FlexWorld can generate high-fidelity scenes with flexible views like 360{\deg} rotations and zooming. Project page: https://ml-gsai.github.io/FlexWorld.