Convex Hull-based Algebraic Constraint for Visual Quadric SLAM

Using Quadrics as the object representation has the benefits of both generality and closed-form projection derivation between image and world spaces. Although numerous constraints have been proposed for dual quadric reconstruction, we found that many of them are imprecise and provide minimal improvements tothis http URLscrutinizing the existing constraints, we introduce a concise yet more precise convex hull-based algebraic constraint for object landmarks, which is applied to object reconstruction, frontend pose estimation, and backend bundlethis http URLconstraint is designed to fully leverage precise semantic segmentation, effectively mitigating mismatches between complex-shaped object contours and dualthis http URLon public datasets demonstrate that our approach is applicable to both monocular and RGB-D SLAM and achieves improved object mapping and localization than existing quadric SLAM methods. The implementation of our method is available atthis https URL.

@article{yu2025_2503.01254,
  title={ Convex Hull-based Algebraic Constraint for Visual Quadric SLAM },
  author={ Xiaolong Yu and Junqiao Zhao and Shuangfu Song and Zhongyang Zhu and Zihan Yuan and Chen Ye and Tiantian Feng },
  journal={arXiv preprint arXiv:2503.01254},
  year={ 2025 }
}