Hi everyone, excited to share our new paper on Plug-and-Play restoration using Spherical Priors.</p>\n<p>The main idea is to replace the usual denoiser/diffusion prior in PnP restoration with a Spherical Encoder prior. SP³ returns a natural-looking image after each iteration, and subsequent iterations improve the quality and consistency. This \"anytime\" restoration behavior is a key advantage over diffusion-based methods that require choosing the number of denoising steps in advance, and running all the way to the end.</p>\n<p>In our experiments, SP³ reaches perceptual quality comparable to zero-shot diffusion and flow priors while being 3–630× faster across restoration tasks.</p>\n<p>Happy to answer questions about the algorithm, results, or the sphere-prior design.</p>\n","updatedAt":"2026-06-16T08:15:45.819Z","author":{"_id":"631321910b24eab4746a27c6","avatarUrl":"/avatars/e43bc701ca53a5f19f0aaa4a0e1e059f.svg","fullname":"Sean Man","name":"seanman","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.8888540863990784},"editors":["seanman"],"editorAvatarUrls":["/avatars/e43bc701ca53a5f19f0aaa4a0e1e059f.svg"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2606.16396","authors":[{"_id":"6a31022aa0d4daae4286034c","user":{"_id":"631321910b24eab4746a27c6","avatarUrl":"/avatars/e43bc701ca53a5f19f0aaa4a0e1e059f.svg","isPro":false,"fullname":"Sean Man","user":"seanman","type":"user","name":"seanman"},"name":"Sean Man","status":"claimed_verified","statusLastChangedAt":"2026-06-16T09:47:21.558Z","hidden":false},{"_id":"6a31022aa0d4daae4286034d","name":"Ron Raphaeli","hidden":false},{"_id":"6a31022aa0d4daae4286034e","user":{"_id":"63511e3aa8822aadf5720ff9","avatarUrl":"/avatars/5798938f1c90d642b2d492b67c77aa4c.svg","isPro":false,"fullname":"Matan Kleiner","user":"matankleiner","type":"user","name":"matankleiner"},"name":"Matan Kleiner","status":"claimed_verified","statusLastChangedAt":"2026-06-16T12:05:35.238Z","hidden":false},{"_id":"6a31022aa0d4daae4286034f","name":"Or Ronai","hidden":false}],"mediaUrls":["https://cdn-uploads.huggingface.co/production/uploads/631321910b24eab4746a27c6/UzZN2IfoeGfyK8vmPDtTY.jpeg","https://cdn-uploads.huggingface.co/production/uploads/631321910b24eab4746a27c6/I6iTqrJi6lt7Q-ngfrJbf.jpeg"],"publishedAt":"2026-06-15T00:00:00.000Z","submittedOnDailyAt":"2026-06-16T00:00:00.000Z","title":"SP^3: Spherical Priors for Plug-and-Play Restoration","submittedOnDailyBy":{"_id":"631321910b24eab4746a27c6","avatarUrl":"/avatars/e43bc701ca53a5f19f0aaa4a0e1e059f.svg","isPro":false,"fullname":"Sean Man","user":"seanman","type":"user","name":"seanman"},"summary":"In this paper, we introduce SP^3, a novel Plug-and-Play algorithm that accelerates maximum a posteriori image restoration by replacing denoisers with Spherical Encoders (SE) as generative priors. SP^3 approximates the intractable proximal prior step by utilizing the SE tightly structured latent space as a robust projection onto the natural image manifold. Alternating this projection with a closed-form data-consistency step, via Half-Quadratic Splitting, achieves stable convergence without requiring gradient computation during inference. This unique formulation unlocks \"anytime\" restoration capabilities, producing sharp, plausible images from the first iteration. Evaluations across a variety of image restoration tasks demonstrate that SP^3 achieves perceptual quality comparable to state-of-the-art zero-shot diffusion and flow methods while being 3-630times faster.","upvotes":8,"discussionId":"6a31022aa0d4daae42860350","projectPage":"https://man-sean.github.io/sp3-website/","githubRepo":"https://github.com/man-sean/SP3","githubRepoAddedBy":"user","ai_summary":"SP³ uses spherical encoders as generative priors to accelerate maximum a posteriori image restoration, enabling fast convergence and high-quality results through structured latent space projections.","ai_keywords":["Plug-and-Play","maximum a posteriori","denoisers","Spherical Encoders","generative priors","latent space","natural image manifold","Half-Quadratic Splitting","stable convergence","anytime restoration","perceptual quality","zero-shot diffusion","flow methods"],"ai_summary_model":"Qwen/Qwen2.5-Coder-32B-Instruct","githubStars":0},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"631321910b24eab4746a27c6","avatarUrl":"/avatars/e43bc701ca53a5f19f0aaa4a0e1e059f.svg","isPro":false,"fullname":"Sean Man","user":"seanman","type":"user"},{"_id":"6317925933a4e2b5d3adafac","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6317925933a4e2b5d3adafac/sgnq_ObPLqx2DciRI43f0.jpeg","isPro":false,"fullname":"ron raphaeli","user":"ronraphaeli","type":"user"},{"_id":"68a42b1971d0c7f03b32f24e","avatarUrl":"/avatars/f65528adea688985ae6e6cbd99ce6eb1.svg","isPro":false,"fullname":"Ron Raphaeli","user":"RonRaphaeliNvidia","type":"user"},{"_id":"63f6497eb29015adc33c1da7","avatarUrl":"/avatars/9119dffcec465b42c490d0e893699a60.svg","isPro":false,"fullname":"Vladimir Kulikov","user":"fallenshock","type":"user"},{"_id":"63f6454a9cbd673030295df4","avatarUrl":"/avatars/90c0e4a1c5dffdabbf2492853741d0d9.svg","isPro":false,"fullname":"Hila Manor","user":"hilamanor","type":"user"},{"_id":"62f0cb39671fc964b5063aba","avatarUrl":"/avatars/02404cd78c395331b42500dd6ced35eb.svg","isPro":false,"fullname":"Roy Ganz","user":"proy","type":"user"},{"_id":"63511e3aa8822aadf5720ff9","avatarUrl":"/avatars/5798938f1c90d642b2d492b67c77aa4c.svg","isPro":false,"fullname":"Matan Kleiner","user":"matankleiner","type":"user"},{"_id":"6a313b0835d89a188bbc348a","avatarUrl":"/avatars/931f77674d6988bae3f03330eac6c7d8.svg","isPro":false,"fullname":"Niv Nayman","user":"NivNayman","type":"user"}],"acceptLanguages":["en"],"dailyPaperRank":0,"markdownContentUrl":"https://huggingface.co/buckets/huggingchat/papers-content/resolve/2606/2606.16396.md","query":{}}">
SP^3: Spherical Priors for Plug-and-Play Restoration
Abstract
SP³ uses spherical encoders as generative priors to accelerate maximum a posteriori image restoration, enabling fast convergence and high-quality results through structured latent space projections.
In this paper, we introduce SP^3, a novel Plug-and-Play algorithm that accelerates maximum a posteriori image restoration by replacing denoisers with Spherical Encoders (SE) as generative priors. SP^3 approximates the intractable proximal prior step by utilizing the SE tightly structured latent space as a robust projection onto the natural image manifold. Alternating this projection with a closed-form data-consistency step, via Half-Quadratic Splitting, achieves stable convergence without requiring gradient computation during inference. This unique formulation unlocks "anytime" restoration capabilities, producing sharp, plausible images from the first iteration. Evaluations across a variety of image restoration tasks demonstrate that SP^3 achieves perceptual quality comparable to state-of-the-art zero-shot diffusion and flow methods while being 3-630times faster.
Community
Hi everyone, excited to share our new paper on Plug-and-Play restoration using Spherical Priors.
The main idea is to replace the usual denoiser/diffusion prior in PnP restoration with a Spherical Encoder prior. SP³ returns a natural-looking image after each iteration, and subsequent iterations improve the quality and consistency. This "anytime" restoration behavior is a key advantage over diffusion-based methods that require choosing the number of denoising steps in advance, and running all the way to the end.
In our experiments, SP³ reaches perceptual quality comparable to zero-shot diffusion and flow priors while being 3–630× faster across restoration tasks.
Happy to answer questions about the algorithm, results, or the sphere-prior design.
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/2606.16396 in a dataset README.md to link it from this page.
Cite arxiv.org/abs/2606.16396 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.