Release the paper on arXiv! Check out DBL-Diffusion’s amazing layered video generation capabilities🔥 </p>\n","updatedAt":"2026-07-30T01:36:48.005Z","author":{"_id":"64d0757c73174cecdfce1247","avatarUrl":"/avatars/7c9d64d47df8c17ba94db5010844f89a.svg","fullname":"KyujinHan","name":"kyujinpy","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":58,"isUserFollowing":false}},"numEdits":1,"identifiedLanguage":{"language":"en","probability":0.7016671299934387},"editors":["kyujinpy"],"editorAvatarUrls":["/avatars/7c9d64d47df8c17ba94db5010844f89a.svg"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2607.25802","authors":[{"_id":"6a6988069d3a1231d492b8f2","user":{"_id":"64d0757c73174cecdfce1247","avatarUrl":"/avatars/7c9d64d47df8c17ba94db5010844f89a.svg","isPro":false,"fullname":"KyujinHan","user":"kyujinpy","type":"user","name":"kyujinpy"},"name":"Kyujin Han","status":"claimed_verified","statusLastChangedAt":"2026-07-29T08:45:04.520Z","hidden":false},{"_id":"6a6988069d3a1231d492b8f3","name":"Seungjoo Shin","hidden":false},{"_id":"6a6988069d3a1231d492b8f4","name":"Sunghyun Cho","hidden":false}],"publishedAt":"2026-07-28T00:00:00.000Z","submittedOnDailyAt":"2026-07-30T00:00:00.000Z","title":"Explicit Layer Modeling for Video Object Insertion and Layer Decomposition","submittedOnDailyBy":{"_id":"64d0757c73174cecdfce1247","avatarUrl":"/avatars/7c9d64d47df8c17ba94db5010844f89a.svg","isPro":false,"fullname":"KyujinHan","user":"kyujinpy","type":"user","name":"kyujinpy"},"summary":"Most video editing systems still lack explicit layered video representations, limiting their ability to perform realistic compositing, object reuse, and consistent manipulation. This limitation is especially pronounced in video object insertion and video layer decomposition, where existing methods rely on implicit inference or per-scene optimization due to the absence of explicit foreground-layer supervision. We introduce TriLayer, a large-scale triplet video dataset containing aligned composite, background, and foreground videos, where the foreground layers include both object appearance and associated visual effects. This explicit supervision enables models to learn layered video representations directly rather than inferring them implicitly. Building on this dataset, we propose DBL-Diffusion, a dual-branch diffusion framework that jointly models RGB composites and RGBA foreground layers through shared denoising and cross-branch interaction. We instantiate the framework in two tasks: DBL-Insert for layered object insertion, which generates explicit RGBA layers for realistic compositing and flexible post-editing, and DBL-Decompose for video layer decomposition, which recovers foreground and background layers using triplet supervision. Experiments demonstrate that explicit layer modeling substantially improves both insertion fidelity and decomposition quality.","upvotes":4,"discussionId":"6a6988069d3a1231d492b8f5","githubRepo":"https://github.com/KyujinHan/DBL-Diffusion","githubRepoAddedBy":"user","githubStars":2,"organization":{"_id":"6a1d7b2c2c850ce78fcd590a","name":"postech-cglab","fullname":"POSTECH Computer Graphics Lab","avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/65ad0940d63812c33e033124/v90h7lJrY0VAWqHLkMcxM.png"}},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"64d0757c73174cecdfce1247","avatarUrl":"/avatars/7c9d64d47df8c17ba94db5010844f89a.svg","isPro":false,"fullname":"KyujinHan","user":"kyujinpy","type":"user"},{"_id":"69bd08309044732b67e1d341","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/noauth/QaSIDNh5GsDhuslXstbuN.png","isPro":false,"fullname":"罗 宇明","user":"alexandernelson","type":"user"},{"_id":"67e668d6b3e3ed96c56852e5","avatarUrl":"/avatars/b860358852a0c4f6a55f2da67da23457.svg","isPro":false,"fullname":"Daeyeon Kim","user":"dkim011006","type":"user"},{"_id":"63c5d43ae2804cb2407e4d43","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1673909278097-noauth.png","isPro":false,"fullname":"xziayro","user":"xziayro","type":"user"}],"acceptLanguages":["en"],"dailyPaperRank":0,"organization":{"_id":"6a1d7b2c2c850ce78fcd590a","name":"postech-cglab","fullname":"POSTECH Computer Graphics Lab","avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/65ad0940d63812c33e033124/v90h7lJrY0VAWqHLkMcxM.png"},"markdownContentUrl":"https://huggingface.co/buckets/huggingchat/papers-content/resolve/2607/2607.25802.md","query":{}}">
Explicit Layer Modeling for Video Object Insertion and Layer Decomposition
Abstract
Most video editing systems still lack explicit layered video representations, limiting their ability to perform realistic compositing, object reuse, and consistent manipulation. This limitation is especially pronounced in video object insertion and video layer decomposition, where existing methods rely on implicit inference or per-scene optimization due to the absence of explicit foreground-layer supervision. We introduce TriLayer, a large-scale triplet video dataset containing aligned composite, background, and foreground videos, where the foreground layers include both object appearance and associated visual effects. This explicit supervision enables models to learn layered video representations directly rather than inferring them implicitly. Building on this dataset, we propose DBL-Diffusion, a dual-branch diffusion framework that jointly models RGB composites and RGBA foreground layers through shared denoising and cross-branch interaction. We instantiate the framework in two tasks: DBL-Insert for layered object insertion, which generates explicit RGBA layers for realistic compositing and flexible post-editing, and DBL-Decompose for video layer decomposition, which recovers foreground and background layers using triplet supervision. Experiments demonstrate that explicit layer modeling substantially improves both insertion fidelity and decomposition quality.
Community
Release the paper on arXiv! Check out DBL-Diffusion’s amazing layered video generation capabilities🔥
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/2607.25802 in a model README.md to link it from this page.
Cite arxiv.org/abs/2607.25802 in a dataset README.md to link it from this page.
Cite arxiv.org/abs/2607.25802 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.