{"ID":2885601,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2508.04147","arxiv_id":"2508.04147","title":"IDCNet: Guided Video Diffusion for Metric-Consistent RGBD Scene Generation with Precise Camera Control","abstract":"We present IDC-Net (Image-Depth Consistency Network), a novel framework designed to generate RGB-D video sequences under explicit camera trajectory control. Unlike approaches that treat RGB and depth generation separately, IDC-Net jointly synthesizes both RGB images and corresponding depth maps within a unified geometry-aware diffusion model. The joint learning framework strengthens spatial and geometric alignment across frames, enabling more precise camera control in the generated sequences. To support the training of this camera-conditioned model and ensure high geometric fidelity, we construct a camera-image-depth consistent dataset with metric-aligned RGB videos, depth maps, and accurate camera poses, which provides precise geometric supervision with notably improved inter-frame geometric consistency. Moreover, we introduce a geometry-aware transformer block that enables fine-grained camera control, enhancing control over the generated sequences. Extensive experiments show that IDC-Net achieves improvements over state-of-the-art approaches in both visual quality and geometric consistency of generated scene sequences. Notably, the generated RGB-D sequences can be directly feed for downstream 3D Scene reconstruction tasks without extra post-processing steps, showcasing the practical benefits of our joint learning framework. See more at https://idcnet-scene.github.io.","short_abstract":"We present IDC-Net (Image-Depth Consistency Network), a novel framework designed to generate RGB-D video sequences under explicit camera trajectory control. Unlike approaches that treat RGB and depth generation separately, IDC-Net jointly synthesizes both RGB images and corresponding depth maps within a unified geometr...","url_abs":"https://arxiv.org/abs/2508.04147","url_pdf":"https://arxiv.org/pdf/2508.04147v1","authors":"[\"Lijuan Liu\",\"Wenfa Li\",\"Dongbo Zhang\",\"Shuo Wang\",\"Shaohui Jiao\"]","published":"2025-08-06T07:19:16Z","proceeding":"cs.CV","tasks":"[\"cs.CV\"]","methods":"[\"Diffusion Model\",\"Transformer\"]","has_code":false}