{"ID":2881352,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2508.12216","arxiv_id":"2508.12216","title":"Splat Feature Solver","abstract":"Feature lifting has emerged as a crucial component in 3D scene understanding, enabling the attachment of rich image feature descriptors (e.g., DINO, CLIP) onto splat-based 3D representations. The core challenge lies in optimally assigning rich general attributes to 3D primitives while addressing the inconsistency issues from multi-view images. We present a unified, kernel- and feature-agnostic formulation of the feature lifting problem as a sparse linear inverse problem, which can be solved efficiently in closed form. Our approach admits a provable upper bound on the global optimal error under convex losses for delivering high quality lifted features. To address inconsistencies and noise in multi-view observations, we introduce two complementary regularization strategies to stabilize the solution and enhance semantic fidelity. Tikhonov Guidance enforces numerical stability through soft diagonal dominance, while Post-Lifting Aggregation filters noisy inputs via feature clustering. Extensive experiments demonstrate that our approach achieves state-of-the-art performance on open-vocabulary 3D segmentation benchmarks, outperforming training-based, grouping-based, and heuristic-forward baselines while producing lifted features in minutes. Our \\textbf{code} is available in the \\href{https://github.com/saliteta/splat-distiller/tree/main}{\\textcolor{blue}{GitHub}}. We provide additional \\href{https://splat-distiller.pages.dev/}{\\textcolor{blue}{website}} for more visualization, as well as the \\href{https://www.youtube.com/watch?v=CH-G5hbvArM}{\\textcolor{blue}{video}}.","short_abstract":"Feature lifting has emerged as a crucial component in 3D scene understanding, enabling the attachment of rich image feature descriptors (e.g., DINO, CLIP) onto splat-based 3D representations. The core challenge lies in optimally assigning rich general attributes to 3D primitives while addressing the inconsistency issue...","url_abs":"https://arxiv.org/abs/2508.12216","url_pdf":"https://arxiv.org/pdf/2508.12216v2","authors":"[\"Butian Xiong\",\"Rong Liu\",\"Kenneth Xu\",\"Meida Chen\",\"Andrew Feng\"]","published":"2025-08-17T03:13:06Z","proceeding":"cs.CV","tasks":"[\"cs.CV\"]","methods":"[]","project_urls":"[\"https://splat-distiller.pages.dev/\"]","has_code":false,"code_links":[{"ID":610802,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_id":2881352,"paper_url":"https://arxiv.org/abs/2508.12216","paper_title":"Splat Feature Solver","repo_url":"https://github.com/saliteta/splat-distiller","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"github_stars":0},{"ID":610803,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_id":2881352,"paper_url":"https://arxiv.org/abs/2508.12216","paper_title":"Splat Feature Solver","repo_url":"https://github.com/RongLiu-Leo/arxsite","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"github_stars":0}]}