{"ID":2890441,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2507.19141","arxiv_id":"2507.19141","title":"DASH: 4D Hash Encoding with Self-Supervised Decomposition for Real-Time Dynamic Scene Rendering","abstract":"Dynamic scene reconstruction is a long-term challenge in 3D vision. Existing plane-based methods in dynamic Gaussian splatting suffer from an unsuitable low-rank assumption, causing feature overlap and poor rendering quality. Although 4D hash encoding provides an explicit representation without low-rank constraints, directly applying it to the entire dynamic scene leads to substantial hash collisions and redundancy. To address these challenges, we present DASH, a real-time dynamic scene rendering framework that employs 4D hash encoding coupled with self-supervised decomposition. Our approach begins with a self-supervised decomposition mechanism that separates dynamic and static components without manual annotations or precomputed masks. Next, we introduce a multiresolution 4D hash encoder for dynamic elements, providing an explicit representation that avoids the low-rank assumption. Finally, we present a spatio-temporal smoothness regularization strategy to mitigate unstable deformation artifacts. Experiments on real-world datasets demonstrate that DASH achieves state-of-the-art dynamic rendering performance, exhibiting enhanced visual quality at real-time speeds of 264 FPS on a single 4090 GPU. Code: https://github.com/chenj02/DASH.","short_abstract":"Dynamic scene reconstruction is a long-term challenge in 3D vision. Existing plane-based methods in dynamic Gaussian splatting suffer from an unsuitable low-rank assumption, causing feature overlap and poor rendering quality. Although 4D hash encoding provides an explicit representation without low-rank constraints, di...","url_abs":"https://arxiv.org/abs/2507.19141","url_pdf":"https://arxiv.org/pdf/2507.19141v2","authors":"[\"Jie Chen\",\"Zhangchi Hu\",\"Peixi Wu\",\"Huyue Zhu\",\"Hebei Li\",\"Xiaoyan Sun\"]","published":"2025-07-25T10:26:01Z","proceeding":"cs.CV","tasks":"[\"cs.CV\"]","methods":"[]","has_code":false,"code_links":[{"ID":611781,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_id":2890441,"paper_url":"https://arxiv.org/abs/2507.19141","paper_title":"DASH: 4D Hash Encoding with Self-Supervised Decomposition for Real-Time Dynamic Scene Rendering","repo_url":"https://github.com/chenj02/DASH","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"github_stars":0}]}