arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2103.04216cs.CV
收藏

虚拟法线:加强几何约束以实现精确和鲁棒的深度预测

Virtual Normal: Enforcing Geometric Constraints for Accurate and Robust Depth Prediction

The University of Adelaide(阿德莱德大学);Monash University(莫纳什大学)

Wei Yin, Yifan Liu, Chunhua Shen

摘要:

单目深度预测在理解3D场景几何中起着至关重要的作用。虽然最近的方法已经取得了令人印象深刻的进展,如像素相对误差的评价指标,大多数方法忽略了在3D空间的几何约束。在这项工作中,我们展示了高阶3D几何约束对深度预测的重要性。通过设计一个损失项,强制执行一个简单的几何约束,即,虚拟法线方向确定随机采样的三个点在重建的三维空间,我们显着提高单目深度估计的准确性和鲁棒性。值得注意的是,虚拟法线损失不仅可以提高学习度量深度的性能,而且可以解开尺度信息,并以更好的形状信息丰富模型。因此,当无法访问绝对度量深度训练数据时,我们可以使用虚拟法线来学习在不同场景上生成的鲁棒仿射不变深度。在实验中,我们展示了在NYU Depth-V2和KITTI上学习度量深度的最新结果。根据高质量的预测深度,我们现在能够直接恢复场景的良好3D结构,例如点云和表面法线,消除了以前依赖其他模型的必要性。为了证明在具有虚拟法线损失的不同数据上学习仿射不变深度的良好通用性,我们构建了一个用于训练仿射不变深度的大规模和不同的数据集,称为Diverse Scene Depth数据集(DiverseDepth),并使用zero-shot测试设置在五个数据集上进行测试。代码可在:https://git.io/Depth

英文摘要:

Monocular depth prediction plays a crucial role in understanding 3D scene geometry. Although recent methods have achieved impressive progress in terms of evaluation metrics such as the pixel-wise relative error, most methods neglect the geometric constraints in the 3D space. In this work, we show the importance of the high-order 3D geometric constraints for depth prediction. By designing a loss term that enforces a simple geometric constraint, namely, virtual normal directions determined by randomly sampled three points in the reconstructed 3D space, we significantly improve the accuracy and robustness of monocular depth estimation. Significantly, the virtual normal loss can not only improve the performance of learning metric depth, but also disentangle the scale information and enrich the model with better shape information. Therefore, when not having access to absolute metric depth training data, we can use virtual normal to learn a robust affine-invariant depth generated on diverse scenes. In experiments, We show state-of-the-art results of learning metric depth on NYU Depth-V2 and KITTI. From the high-quality predicted depth, we are now able to recover good 3D structures of the scene such as the point cloud and surface normal directly, eliminating the necessity of relying on additional models as was previously done. To demonstrate the excellent generalizability of learning affine-invariant depth on diverse data with the virtual normal loss, we construct a large-scale and diverse dataset for training affine-invariant depth, termed Diverse Scene Depth dataset (DiverseDepth), and test on five datasets with the zero-shot test setting. Code is available at: https://git.io/Depth

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑