[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-86074-en":3,"doc-seo-86074-105":30,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":21,"is_downloadable":21,"audit_status":21,"page_count":22,"language":23,"language_code":24,"site_id":25,"html_lang":24,"table_of_contents":26,"faqs":27,"seo_title":13,"seo_description":14,"update_tm":28,"read_time":29},86074,2336464648746,"Skyler","https://ap-avatar.wpscdn.com/davatar_276721f389ce27ea32af1340a28f341c",8,"Research & Report","DP-Splat Bayesian Nonparametric Complexity Control for Gaussian Splatting","3D Gaussian Splatting models scenes as a finite mixture of anisotropic Gaussians, where the component count K is usually controlled by heuristics or a user cap. Variational Bayes Gaussian Splatting fixes K and limits uncertainty to per-component effects. DP-Splat replaces the finite Dirichlet mixture prior with a truncated stick-breaking Dirichlet-process prior, and also provides a sparse overfitted finite Dirichlet alternative. The method preserves closed-form CAVI updates, offers monotonicity, a corrected truncation-error bound, and evaluates complexity selection versus per-component efficiency.","arXiv :2607 . 10912v1 [ cs .CV] 12 Jul 2026  \nDP-Splat: Bayesian Nonparametric Complexity Control for Gaussian Splatting  \nAqi Dong  \nEmbry-Riddle Aeronautical University  \n[donga2@erau.edu](donga2@erau.edu)  \nAbstract  \n3D Gaussian Splatting represents scenes as finite mixtures of anisotropic Gaussians whose number of components K is governed by heuristic adaptive density control or user-specified caps. Variational Bayes Gaussian Splatting (VBGS) recast splat fitting as conjugate variational inference over a finite mixture, but K remains fixed. We replace the finite symmetric Dirichlet over mixture weights with a truncated stick-breaking Dirichlet-process prior—and, as a theory-backed alternative, a sparse overfitted finite Dirichlet—so that the number of occupied components adapts to the data while every update remains a closed-form coordinateascent step (Beta/Categorical/Normal–Inverse–Wishart); a natural-gradient stochastic variant makes the per-step cost independent of the number of points. We give an exact monotonicity guarantee, a rigorous truncation-error bound that corrects an anti-conservative large-α approximation in common use, and an honest account of what the fitted number of componto sceneentscomdpoeslexiatnydadnodersenocotveestimrs theatetru. EemKpi(ricallywithin, )1)thoeneffweecllt-ivesepcaompleratedxsyityntiacdadptatsa with regime-appropriate concentration; (ii) a deconfounded three-way image comparison shows the DP prior’s contribution is complexity selection, not per-component efficiency:  \nconverged DP fits exceed single-pass fixed-K VBGS at matched budgets by +2 .7 dB on average yet tie an equally converged fixed-K baseline, while on 3D scenes DP-Splat matches or exceeds VBGS’s held-out color prediction with 5.9–7.6 × fewer components (single-pass fits at compute comparable to VBGS’s already match with ≈5× fewer); (iii) the posteriorpredictive color variance is well calibrated on model-matched synthetic data (regression-ECE ≈ 3 × 10 −3); and (iv) the ordering suggested by exact-posterior asymptotics reverses under mean-field coordinate ascent at practical N: the Dirichlet-process prior resists over-splitting while the sparse finite mixture saturates its truncation—a gap between variational practice and posterior asymptotics that we document across three orders of magnitude in N. Code, exact reproduction commands, and all experiment records accompany the submission.  \n1 Introduction  \n3D Gaussian Splatting (3DGS) (Kerbl et al., 2023) represents a scene as a mixture of anisotropic 3D Gaussiansand renders it with a differentiable rasterizer, achieving state-of-the-art novel-view synthesis at real-timeframe rates. A central but under-examined design choice is the number of components: 3DGS grows and prunes Gaussians with adaptive density control—gradient-magnitude splitting and opacity thresholds—that works well in practice but is a stack of heuristics with tunable knobs and no statistical semantics. Follow-up work has made the dynamics more principled (e.g., MCMC-style relocation moves; Kheradmand et al., 2024) yet still requires a user-specified cap on the component count.  \nVariational Bayes Gaussian Splatting (VBGS; Van de Maele et al., 2024) reframed splat fitting as conjugate variational inference over a finite mixture on colored points: closed-form coordinate-ascent variational inference (CAVI) replaces stochastic gradients, enabling continual learning without replay. But the mixture is finite with K fixed in advance, uncertainty is per-component only, and VBGS inherits a component-recycling heuristic to revive dead components—adaptive density control through the back door.  \nIn the statistics literature, letting a mixture choose its own complexity is a mature problem with three standard routes: Dirichlet-process (DP) mixtures with truncated variational inference (Ferguson, 1973; Sethuraman, 1994; Ishwaran & James, 2001; Blei & Jordan, 2006); sparse overfitted finite mixtures whose superfluous componen","cbCaicdv4y4riNd5","https://ap.wps.com/l/cbCaicdv4y4riNd5","pdf",1948247,4,1,18,"English","en",105,"# Introduction\n# DP-Splat Method\n## Nonparametric priors and closed-form updates\n## Truncation and error bounds\n## Experiments and comparisons","[{\"question\":\"What problem does DP-Splat address in Gaussian Splatting?\",\"answer\":\"DP-Splat addresses the lack of statistical semantics for choosing the number of Gaussian components K in 3D Gaussian Splatting, which is typically controlled by heuristics or fixed caps.\"},{\"question\":\"How does DP-Splat make the component count adapt to data?\",\"answer\":\"DP-Splat replaces the fixed finite Dirichlet mixture weights with a truncated stick-breaking Dirichlet-process prior (and also considers a sparse overfitted finite Dirichlet), allowing occupied components to adapt to the data while updates remain conjugate and closed-form.\"},{\"question\":\"What theoretical assurances and evaluation outcomes are reported?\",\"answer\":\"The paper provides an exact monotonicity guarantee and a rigorous truncation-error bound that corrects a commonly used approximation. Experiments report improved or matched performance with fewer components and calibrated posterior-predictive color variance.\"}]",1784208354,45,{"code":4,"msg":31,"data":32},"ok",{"site_id":25,"language":24,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"dp-splat-bayesian-nonparametric-complexity-control-for-gaussian-splatting","",{"@graph":36,"@context":85},[37,53,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":21},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":20},"https://docshare.wps.com/document/dp-splat-bayesian-nonparametric-complexity-control-for-gaussian-splatting/86074/",{"url":52,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":24,"description":14,"dateModified":61,"datePublished":62,"encodingFormat":60,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":41,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-07-27","2026-07-16",true,{"@type":65,"interactionType":66,"userInteractionCount":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"What problem does DP-Splat address in Gaussian Splatting?","Question",{"text":75,"@type":76},"DP-Splat addresses the lack of statistical semantics for choosing the number of Gaussian components K in 3D Gaussian Splatting, which is typically controlled by heuristics or fixed caps.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does DP-Splat make the component count adapt to data?",{"text":80,"@type":76},"DP-Splat replaces the fixed finite Dirichlet mixture weights with a truncated stick-breaking Dirichlet-process prior (and also considers a sparse overfitted finite Dirichlet), allowing occupied components to adapt to the data while updates remain conjugate and closed-form.",{"name":82,"@type":73,"acceptedAnswer":83},"What theoretical assurances and evaluation outcomes are reported?",{"text":84,"@type":76},"The paper provides an exact monotonicity guarantee and a rigorous truncation-error bound that corrects a commonly used approximation. Experiments report improved or matched performance with fewer components and calibrated posterior-predictive color variance.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":25},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,135],{"id":21,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":20,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]