{"ID":2832595,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2512.05783","arxiv_id":"2512.05783","title":"Curvature-Regularized Variational Autoencoder for 3D Scene Reconstruction from Sparse Depth","abstract":"When depth sensors provide only 5% of needed measurements, reconstructing complete 3D scenes becomes difficult. Autonomous vehicles and robots cannot tolerate the geometric errors that sparse reconstruction introduces. We propose curvature regularization through a discrete Laplacian operator, achieving 18.1% better reconstruction accuracy than standard variational autoencoders. Our contribution challenges an implicit assumption in geometric deep learning: that combining multiple geometric constraints improves performance. A single well-designed regularization term not only matches but exceeds the effectiveness of complex multi-term formulations. The discrete Laplacian offers stable gradients and noise suppression with just 15% training overhead and zero inference cost. Code and models are available at https://github.com/Maryousefi/GeoVAE-3D.","short_abstract":"When depth sensors provide only 5% of needed measurements, reconstructing complete 3D scenes becomes difficult. Autonomous vehicles and robots cannot tolerate the geometric errors that sparse reconstruction introduces. We propose curvature regularization through a discrete Laplacian operator, achieving 18.1% better rec...","url_abs":"https://arxiv.org/abs/2512.05783","url_pdf":"https://arxiv.org/pdf/2512.05783v1","authors":"[\"Maryam Yousefi\",\"Soodeh Bakhshandeh\"]","published":"2025-12-05T15:11:04Z","proceeding":"cs.CV","tasks":"[\"cs.CV\",\"cs.LG\"]","methods":"[\"Variational Autoencoder\"]","has_code":false,"code_links":[{"ID":606259,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_id":2832595,"paper_url":"https://arxiv.org/abs/2512.05783","paper_title":"Curvature-Regularized Variational Autoencoder for 3D Scene Reconstruction from Sparse Depth","repo_url":"https://github.com/Maryousefi/GeoVAE-3D","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"github_stars":0}]}