Tech report.<br>ABot-Earth-0.5 is a generative 3D Earth model developed by the Amap CV Lab.<br>The github repo is dedicated to our technical report and academic discourse. It does not contain implementation code.<br>Media: <a href=\"https://www.youtube.com/watch?v=Qv64FpsEi2Q\" rel=\"nofollow\">https://www.youtube.com/watch?v=Qv64FpsEi2Q</a><br>Dive into our live demo! Experience ABot-Earth firsthand on our project page: <a href=\"https://abot-earth.amap.com/\" rel=\"nofollow\">https://abot-earth.amap.com/</a></p>\n","updatedAt":"2026-06-10T08:50:25.783Z","author":{"_id":"6433bc32a4c9c55871a42812","avatarUrl":"/avatars/c27c063d02419b8ba32657514c07e3b2.svg","fullname":"qian#143","name":"qian43","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":6,"isUserFollowing":false}},"numEdits":1,"identifiedLanguage":{"language":"en","probability":0.8614749312400818},"editors":["qian43"],"editorAvatarUrls":["/avatars/c27c063d02419b8ba32657514c07e3b2.svg"],"reactions":[],"isReport":false}},{"id":"6a2911815fe42117b1ad8186","author":{"_id":"65e56ff71db33cf9eccdb1e1","avatarUrl":"/avatars/fad1380c0bf77c58af48c27cf3abaef0.svg","fullname":"Yuting Xu","name":"RainyX","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"isUserFollowing":false},"createdAt":"2026-06-10T07:25:53.000Z","type":"comment","data":{"edited":false,"hidden":false,"latest":{"raw":"Interesting! Nice work!","html":"<p>Interesting! Nice work!</p>\n","updatedAt":"2026-06-10T07:25:53.267Z","author":{"_id":"65e56ff71db33cf9eccdb1e1","avatarUrl":"/avatars/fad1380c0bf77c58af48c27cf3abaef0.svg","fullname":"Yuting Xu","name":"RainyX","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.8135697245597839},"editors":["RainyX"],"editorAvatarUrls":["/avatars/fad1380c0bf77c58af48c27cf3abaef0.svg"],"reactions":[],"isReport":false}},{"id":"6a291308b8c386266039e2f0","author":{"_id":"6985807171cad19e9854ba5e","avatarUrl":"/avatars/204d6d531e69821fe38490c6634eb666.svg","fullname":"xumu","name":"xumu123","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"isUserFollowing":false},"createdAt":"2026-06-10T07:32:24.000Z","type":"comment","data":{"edited":true,"hidden":true,"hiddenBy":"","latest":{"raw":"This comment has been hidden","html":"This comment has been hidden","updatedAt":"2026-06-10T07:32:40.015Z","author":{"_id":"6985807171cad19e9854ba5e","avatarUrl":"/avatars/204d6d531e69821fe38490c6634eb666.svg","fullname":"xumu","name":"xumu123","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"isUserFollowing":false}},"numEdits":0,"editors":[],"editorAvatarUrls":[],"reactions":[]}}],"primaryEmailConfirmed":false,"paper":{"id":"2606.09967","authors":[{"_id":"6a28c3c5e7d78ea7587e52bd","name":"Ming Qian","hidden":false},{"_id":"6a28c3c5e7d78ea7587e52be","name":"Tianjian Ouyang","hidden":false},{"_id":"6a28c3c5e7d78ea7587e52bf","name":"Mingchao Sun","hidden":false},{"_id":"6a28c3c5e7d78ea7587e52c0","name":"Zijian Wang","hidden":false},{"_id":"6a28c3c5e7d78ea7587e52c1","name":"Jincheng Xiong","hidden":false},{"_id":"6a28c3c5e7d78ea7587e52c2","name":"Jiarong Han","hidden":false},{"_id":"6a28c3c5e7d78ea7587e52c3","name":"Yongchang Zhang","hidden":false},{"_id":"6a28c3c5e7d78ea7587e52c4","name":"Jiawei Zhang","hidden":false},{"_id":"6a28c3c5e7d78ea7587e52c5","name":"Xu Wang","hidden":false},{"_id":"6a28c3c5e7d78ea7587e52c6","name":"Yu Liu","hidden":false},{"_id":"6a28c3c5e7d78ea7587e52c7","name":"Luyang Tang","hidden":false},{"_id":"6a28c3c5e7d78ea7587e52c8","name":"Fei Yu","hidden":false},{"_id":"6a28c3c5e7d78ea7587e52c9","name":"Zengye Ge","hidden":false},{"_id":"6a28c3c5e7d78ea7587e52ca","name":"Mengmeng Du","hidden":false},{"_id":"6a28c3c5e7d78ea7587e52cb","name":"Yuan Liu","hidden":false},{"_id":"6a28c3c5e7d78ea7587e52cc","name":"Nianfei Fan","hidden":false},{"_id":"6a28c3c5e7d78ea7587e52cd","name":"Song Wang","hidden":false},{"_id":"6a28c3c5e7d78ea7587e52ce","name":"Yingliang Peng","hidden":false},{"_id":"6a28c3c5e7d78ea7587e52cf","name":"Chunxue Jia","hidden":false},{"_id":"6a28c3c5e7d78ea7587e52d0","name":"Yang Liu","hidden":false},{"_id":"6a28c3c5e7d78ea7587e52d1","name":"Shiying Zeng","hidden":false},{"_id":"6a28c3c5e7d78ea7587e52d2","name":"Haozhe Shi","hidden":false},{"_id":"6a28c3c5e7d78ea7587e52d3","name":"Junnan Lai","hidden":false},{"_id":"6a28c3c5e7d78ea7587e52d4","name":"Hongyu Pan","hidden":false},{"_id":"6a28c3c5e7d78ea7587e52d5","name":"Zheng Wu","hidden":false},{"_id":"6a28c3c5e7d78ea7587e52d6","name":"Ning Guo","hidden":false},{"_id":"6a28c3c5e7d78ea7587e52d7","name":"Mu Xu","hidden":false},{"_id":"6a28c3c5e7d78ea7587e52d8","name":"Hang Zhang","hidden":false}],"publishedAt":"2026-06-08T00:00:00.000Z","submittedOnDailyAt":"2026-06-10T00:00:00.000Z","title":"ABot-Earth 0.5: Generative 3D Earth Model","submittedOnDailyBy":{"_id":"6433bc32a4c9c55871a42812","avatarUrl":"/avatars/c27c063d02419b8ba32657514c07e3b2.svg","isPro":false,"fullname":"qian#143","user":"qian43","type":"user","name":"qian43"},"summary":"We present ABot-Earth 0.5, a generative 3D framework designed to synthesize vast, seamless 3D environments from ubiquitous, geospatially referenced satellite imagery. To achieve this, we propose a novel generative model formulated directly with the 3D Gaussian Splatting (3DGS) representation. The model is trained on a diverse corpus of existing real-world urban reconstructions, learning to generate realistic geometry and textures. At inference, it synthesizes novel 3D scenes conditioned solely on satellite imagery at a scalable rate of under 10 minutes per square kilometer, while demonstrating exceptional realism. The framework is designed for accessibility, with integrated hierarchical level-of-detail (LOD) structures that permit real-time, interactive visualization on web-based map engines. This high-fidelity simulation sandbox effectively mitigates the sim-to-real domain gap, enabling critical downstream Embodied AI applications like closed-loop UAV navigation. By providing an ultra-low-cost and high-efficiency solution, ABot-Earth 0.5 significantly lowers the technical and financial barriers to large-scale 3D reconstruction and empowers the future of global digital earth visualization.","upvotes":193,"discussionId":"6a28c3c5e7d78ea7587e52d9","projectPage":"https://abot-earth.amap.com/","githubRepo":"https://github.com/amap-cvlab/ABot-Earth-0.5","githubRepoAddedBy":"user","ai_summary":"ABot-Earth 0.5 generates realistic 3D environments from satellite imagery using 3D Gaussian Splatting representation, enabling fast synthesis and real-time visualization for Embodied AI applications.","ai_keywords":["3D Gaussian Splatting","generative model","satellite imagery","3D environment synthesis","level-of-detail","real-time visualization","Embodied AI","UAV navigation","digital earth visualization"],"ai_summary_model":"Qwen/Qwen2.5-Coder-32B-Instruct","githubStars":104,"organization":{"_id":"641415d08900ef6afa2fcb73","name":"acvlab","fullname":"Alibaba AMAP CV Lab","avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/6414106ce7d5f817d204e160/dfveRtrRy8Xn7QpG684zl.png"}},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"6433bc32a4c9c55871a42812","avatarUrl":"/avatars/c27c063d02419b8ba32657514c07e3b2.svg","isPro":false,"fullname":"qian#143","user":"qian43","type":"user"},{"_id":"6039478ab3ecf716b1a5fd4d","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6039478ab3ecf716b1a5fd4d/_Thy4E7taiSYBLKxEKJbT.jpeg","isPro":true,"fullname":"taesiri","user":"taesiri","type":"user"},{"_id":"646329191bb15fa38b0a6a21","avatarUrl":"/avatars/6393cad9da0bf5f7722d99b0ef431246.svg","isPro":false,"fullname":"hang zhang","user":"whuhangzhang","type":"user"},{"_id":"67c6a971bd77c0d90dea86d9","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/no-auth/1MVSYolc2ASZ8AF9pZBH9.png","isPro":false,"fullname":"xu","user":"mrxu6","type":"user"},{"_id":"64fa1695b60e2c9ddde44b1c","avatarUrl":"/avatars/a23096bdce27d934c377dde13c449afc.svg","isPro":false,"fullname":"hanhui","user":"clearhanhui","type":"user"},{"_id":"6985807171cad19e9854ba5e","avatarUrl":"/avatars/204d6d531e69821fe38490c6634eb666.svg","isPro":false,"fullname":"xumu","user":"xumu123","type":"user"},{"_id":"62653e93e25d2b490db759d1","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1650802315262-noauth.jpeg","isPro":false,"fullname":"Hengda Xu","user":"DaDaMrX","type":"user"},{"_id":"662b5ee41ce08a97563c3353","avatarUrl":"/avatars/d4dc86b8c79f8f06060aac1136e7a1a0.svg","isPro":false,"fullname":"Shen Jiajun","user":"NOBULON","type":"user"},{"_id":"691d83fc4e75e79982d42086","avatarUrl":"/avatars/8228dd170139247f72a2b4c381ff5942.svg","isPro":false,"fullname":"Zhicheng Liu","user":"jackmiemie","type":"user"},{"_id":"647b2c7b2d27d3541decf7a1","avatarUrl":"/avatars/003ea8e46fec9869876280679c7bbbcf.svg","isPro":false,"fullname":"Zedong Chu","user":"jellyczd","type":"user"},{"_id":"642f83fc4e7b53c5addc2a25","avatarUrl":"/avatars/660d0900ba0771d60d0f3c348803daf9.svg","isPro":false,"fullname":"Young","user":"MarvinYoung","type":"user"},{"_id":"68a74579646ad764c5cafd71","avatarUrl":"/avatars/3cb44f4373d708e5c397dd4814a716eb.svg","isPro":false,"fullname":"MihailSlutsky","user":"MihailSlutsky","type":"user"}],"acceptLanguages":["en"],"dailyPaperRank":1,"organization":{"_id":"641415d08900ef6afa2fcb73","name":"acvlab","fullname":"Alibaba AMAP CV Lab","avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/6414106ce7d5f817d204e160/dfveRtrRy8Xn7QpG684zl.png"},"markdownContentUrl":"https://huggingface.co/buckets/huggingchat/papers-content/resolve/2606/2606.09967.md"}">
ABot-Earth 0.5: Generative 3D Earth Model
Authors: ,
,
,
,
,
,
,
,
,
,
,
,
,
,
,
,
,
,
,
,
,
Abstract
ABot-Earth 0.5 generates realistic 3D environments from satellite imagery using 3D Gaussian Splatting representation, enabling fast synthesis and real-time visualization for Embodied AI applications.
We present ABot-Earth 0.5, a generative 3D framework designed to synthesize vast, seamless 3D environments from ubiquitous, geospatially referenced satellite imagery. To achieve this, we propose a novel generative model formulated directly with the 3D Gaussian Splatting (3DGS) representation. The model is trained on a diverse corpus of existing real-world urban reconstructions, learning to generate realistic geometry and textures. At inference, it synthesizes novel 3D scenes conditioned solely on satellite imagery at a scalable rate of under 10 minutes per square kilometer, while demonstrating exceptional realism. The framework is designed for accessibility, with integrated hierarchical level-of-detail (LOD) structures that permit real-time, interactive visualization on web-based map engines. This high-fidelity simulation sandbox effectively mitigates the sim-to-real domain gap, enabling critical downstream Embodied AI applications like closed-loop UAV navigation. By providing an ultra-low-cost and high-efficiency solution, ABot-Earth 0.5 significantly lowers the technical and financial barriers to large-scale 3D reconstruction and empowers the future of global digital earth visualization.
Community
Tech report.
ABot-Earth-0.5 is a generative 3D Earth model developed by the Amap CV Lab.
The github repo is dedicated to our technical report and academic discourse. It does not contain implementation code.
Media: https://www.youtube.com/watch?v=Qv64FpsEi2Q
Dive into our live demo! Experience ABot-Earth firsthand on our project page: https://abot-earth.amap.com/
This comment has been hidden Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.
Tap or paste here to upload images
Cite arxiv.org/abs/2606.09967 in a model README.md to link it from this page.
Cite arxiv.org/abs/2606.09967 in a dataset README.md to link it from this page.
Cite arxiv.org/abs/2606.09967 in a Space README.md to link it from this page.
Discussion (0)
Sign in to join the discussion. Free account, 30 seconds — email code or GitHub.
Sign in →No comments yet. Sign in and be the first to say something.