{"ID":2824332,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2512.23437","arxiv_id":"2512.23437","title":"RealX3D: A Physically-Degraded 3D Benchmark for Multi-view Visual Restoration and Reconstruction","abstract":"We introduce RealX3D, a real-capture benchmark for multi-view visual restoration and 3D reconstruction under diverse physical degradations. RealX3D groups corruptions into four families, including illumination, scattering, occlusion, and blurring, and captures each at multiple severity levels using a unified acquisition protocol that yields pixel-aligned LQ/GT views. Each scene includes high-resolution capture, RAW images, and dense laser scans, from which we derive world-scale meshes and metric depth. Benchmarking a broad range of optimization-based and feed-forward methods shows substantial degradation in reconstruction quality under physical corruptions, underscoring the fragility of current multi-view pipelines in real-world challenging environments.","short_abstract":"We introduce RealX3D, a real-capture benchmark for multi-view visual restoration and 3D reconstruction under diverse physical degradations. RealX3D groups corruptions into four families, including illumination, scattering, occlusion, and blurring, and captures each at multiple severity levels using a unified acquisitio...","url_abs":"https://arxiv.org/abs/2512.23437","url_pdf":"https://arxiv.org/pdf/2512.23437v2","authors":"[\"Shuhong Liu\",\"Chenyu Bao\",\"Ziteng Cui\",\"Yun Liu\",\"Xuangeng Chu\",\"Lin Gu\",\"Marcos V. Conde\",\"Ryo Umagami\",\"Tomohiro Hashimoto\",\"Zijian Hu\",\"Tianhan Xu\",\"Yuan Gan\",\"Yusuke Kurose\",\"Tatsuya Harada\"]","published":"2025-12-29T12:57:19Z","proceeding":"cs.CV","tasks":"[\"cs.CV\",\"cs.MM\"]","methods":"[]","has_code":false}