{"ID":2827404,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2512.16219","arxiv_id":"2512.16219","title":"Learning High-Quality Initial Noise for Single-View Synthesis with Diffusion Models","abstract":"Single-view novel view synthesis (NVS) models based on diffusion models have recently attracted increasing attention, as they can generate a series of novel view images from a single image prompt and camera pose information as conditions. It has been observed that in diffusion models, certain high-quality initial noise patterns lead to better generation results than others. However, there remains a lack of dedicated learning frameworks that enable NVS models to learn such high-quality noise. To obtain high-quality initial noise from random Gaussian noise, we make the following contributions. First, we design a discretized Euler inversion method to inject image semantic information into random noise, thereby constructing paired datasets of random and high-quality noise. Second, we propose a learning framework based on an encoder-decoder network (EDN) that directly transforms random noise into high-quality noise. Experiments demonstrate that the proposed EDN can be seamlessly plugged into various NVS models, such as SV3D and MV-Adapter, achieving significant performance improvements across multiple datasets. Code is available at: https://github.com/zhihao0512/EDN.","short_abstract":"Single-view novel view synthesis (NVS) models based on diffusion models have recently attracted increasing attention, as they can generate a series of novel view images from a single image prompt and camera pose information as conditions. It has been observed that in diffusion models, certain high-quality initial noise...","url_abs":"https://arxiv.org/abs/2512.16219","url_pdf":"https://arxiv.org/pdf/2512.16219v1","authors":"[\"Zhihao Zhang\",\"Xuejun Yang\",\"Weihua Liu\",\"Mouquan Shen\"]","published":"2025-12-18T06:08:21Z","proceeding":"cs.CV","tasks":"[\"cs.CV\",\"eess.IV\"]","methods":"[\"Diffusion Model\"]","has_code":false,"code_links":[{"ID":605804,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_id":2827404,"paper_url":"https://arxiv.org/abs/2512.16219","paper_title":"Learning High-Quality Initial Noise for Single-View Synthesis with Diffusion Models","repo_url":"https://github.com/zhihao0512/EDN","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"github_stars":0}]}