{"ID":2884392,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2508.06982","arxiv_id":"2508.06982","title":"IntrinsicWeather: Controllable Weather Editing in Intrinsic Space","abstract":"We present IntrinsicWeather, a diffusion-based framework for controllable weather editing in intrinsic space. Our framework includes two components based on diffusion priors: an inverse renderer that estimates material properties, scene geometry, and lighting as intrinsic maps from an input image, and a forward renderer that utilizes these geometry and material maps along with a text prompt that describes specific weather conditions to generate a final image. The intrinsic maps enhance controllability compared to traditional pixel-space editing approaches. We propose an intrinsic map-aware attention mechanism that improves spatial correspondence and decomposition quality in large outdoor scenes. For forward rendering, we leverage CLIP-space interpolation of weather prompts to achieve fine-grained weather control. We also introduce a synthetic and a real-world dataset, containing 38k and 18k images under various weather conditions, each with intrinsic map annotations. IntrinsicWeather outperforms state-of-the-art pixel-space editing approaches, weather restoration methods, and rendering-based methods, showing promise for downstream tasks such as autonomous driving, enhancing the robustness of detection and segmentation in challenging weather scenarios.","short_abstract":"We present IntrinsicWeather, a diffusion-based framework for controllable weather editing in intrinsic space. Our framework includes two components based on diffusion priors: an inverse renderer that estimates material properties, scene geometry, and lighting as intrinsic maps from an input image, and a forward rendere...","url_abs":"https://arxiv.org/abs/2508.06982","url_pdf":"https://arxiv.org/pdf/2508.06982v7","authors":"[\"Yixin Zhu\",\"Zuo-Liang Zhu\",\"Jian Yang\",\"Miloš Hašan\",\"Jin Xie\",\"Beibei Wang\"]","published":"2025-08-09T13:29:39Z","proceeding":"cs.CV","tasks":"[\"cs.CV\",\"cs.AI\"]","methods":"[\"Diffusion Model\"]","has_code":false}