{"ID":2867318,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2509.19297","arxiv_id":"2509.19297","title":"VolSplat: Rethinking Feed-Forward 3D Gaussian Splatting with Voxel-Aligned Prediction","abstract":"Feed-forward 3D Gaussian Splatting (3DGS) has emerged as a highly effective solution for novel view synthesis. Existing methods predominantly rely on a \\emph{pixel-aligned} Gaussian prediction paradigm, where each 2D pixel is mapped to a 3D Gaussian. We rethink this widely adopted formulation and identify several inherent limitations: it renders the reconstructed 3D models heavily dependent on the number of input views, leads to view-biased density distributions, and introduces alignment errors, particularly when source views contain occlusions or low texture. To address these challenges, we introduce VolSplat, a new multi-view feed-forward paradigm that replaces pixel alignment with voxel-aligned Gaussians. By directly predicting Gaussians from a predicted 3D voxel grid, it overcomes pixel alignment's reliance on error-prone 2D feature matching, ensuring robust multi-view consistency. Furthermore, it enables adaptive control over density based on 3D scene complexity, yielding more faithful Gaussians, improved geometric consistency, and enhanced novel-view rendering quality. Experiments on widely used benchmarks demonstrate that VolSplat achieves state-of-the-art performance, while producing more plausible and view-consistent results. The video results, code and trained models are available on our project page: https://lhmd.top/volsplat.","short_abstract":"Feed-forward 3D Gaussian Splatting (3DGS) has emerged as a highly effective solution for novel view synthesis. Existing methods predominantly rely on a \\emph{pixel-aligned} Gaussian prediction paradigm, where each 2D pixel is mapped to a 3D Gaussian. We rethink this widely adopted formulation and identify several inher...","url_abs":"https://arxiv.org/abs/2509.19297","url_pdf":"https://arxiv.org/pdf/2509.19297v2","authors":"[\"Weijie Wang\",\"Yeqing Chen\",\"Zeyu Zhang\",\"Hengyu Liu\",\"Haoxiao Wang\",\"Zhiyuan Feng\",\"Wenkang Qin\",\"Feng Chen\",\"Zheng Zhu\",\"Donny Y. Chen\",\"Bohan Zhuang\"]","published":"2025-09-23T17:59:02Z","proceeding":"cs.CV","tasks":"[\"cs.CV\"]","methods":"[]","project_urls":"[\"https://lhmd.top/volsplat\"]","has_code":false,"code_links":[{"ID":609453,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_id":2867318,"paper_url":"https://arxiv.org/abs/2509.19297","paper_title":"VolSplat: Rethinking Feed-Forward 3D Gaussian Splatting with Voxel-Aligned Prediction","repo_url":"https://github.com/ziplab/VolSplat","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"github_stars":0}]}