{"ID":2833248,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2512.03350","arxiv_id":"2512.03350","title":"SeeU: Seeing the Unseen World via 4D Dynamics-aware Generation","abstract":"Images and videos are discrete 2D projections of the 4D world (3D space + time). Most visual understanding, prediction, and generation operate directly on 2D observations, leading to suboptimal performance. We propose SeeU, a novel approach that learns the continuous 4D dynamics and generate the unseen visual contents. The principle behind SeeU is a new 2D$\\to$4D$\\to$2D learning framework. SeeU first reconstructs the 4D world from sparse and monocular 2D frames (2D$\\to$4D). It then learns the continuous 4D dynamics on a low-rank representation and physical constraints (discrete 4D$\\to$continuous 4D). Finally, SeeU rolls the world forward in time, re-projects it back to 2D at sampled times and viewpoints, and generates unseen regions based on spatial-temporal context awareness (4D$\\to$2D). By modeling dynamics in 4D, SeeU achieves continuous and physically-consistent novel visual generation, demonstrating strong potentials in multiple tasks including unseen temporal generation, unseen spatial generation, and video editing. All data and code will be public at https://yuyuanspace.com/SeeU/","short_abstract":"Images and videos are discrete 2D projections of the 4D world (3D space + time). Most visual understanding, prediction, and generation operate directly on 2D observations, leading to suboptimal performance. We propose SeeU, a novel approach that learns the continuous 4D dynamics and generate the unseen visual contents....","url_abs":"https://arxiv.org/abs/2512.03350","url_pdf":"https://arxiv.org/pdf/2512.03350v2","authors":"[\"Yu Yuan\",\"Tharindu Wickremasinghe\",\"Zeeshan Nadir\",\"Xijun Wang\",\"Yiheng Chi\",\"Stanley H. Chan\"]","published":"2025-12-03T01:30:45Z","proceeding":"cs.CV","tasks":"[\"cs.CV\"]","methods":"[]","project_urls":"[\"https://yuyuanspace.com/SeeU/\"]","has_code":false,"code_links":[{"ID":606307,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_id":2833248,"paper_url":"https://arxiv.org/abs/2512.03350","paper_title":"SeeU: Seeing the Unseen World via 4D Dynamics-aware Generation","repo_url":"https://github.com/pandayuanyu/SeeU","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"github_stars":0},{"ID":606308,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_id":2833248,"paper_url":"https://arxiv.org/abs/2512.03350","paper_title":"SeeU: Seeing the Unseen World via 4D Dynamics-aware Generation","repo_url":"https://github.com/nerfies/nerfies.github.io","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"github_stars":0}]}