SCIPaD：将空间线索融入无监督位姿-深度联合学习
SCIPaD: Incorporating Spatial Clues into Unsupervised Pose-Depth Joint Learning
AI总结：
针对无监督单目深度与位姿联合学习中相机位姿估计不准的问题，本文提出 SCIPaD，利用置信度特征流、位置线索聚合和分层嵌入注入融合空间线索，在 KITTI Odometry 上显著降低位姿误差。
摘要：
无监督单目深度估计框架已在自动驾驶中展现出良好性能。然而，现有方案主要依赖简单的卷积神经网络来恢复自身运动，在动态、复杂的真实场景中难以估计精确的相机位姿。这些不准确的相机位姿不可避免地会恶化光度重建，并以错误的监督信号误导深度估计网络。本文提出 SCIPaD，这是一种将空间线索融入无监督深度-位姿联合学习的新颖方法。具体而言，提出了一种置信度感知的特征流估计器，用于获取二维特征位置平移及其相关置信度。同时，引入位置线索聚合器，将来自 DepthNet 的伪三维点云与二维特征流整合为齐次位置表示。最后，提出分层位置嵌入注入器，用于有选择地将空间线索注入语义特征，以实现鲁棒的相机位姿解码。大量实验和分析表明，与其他最先进方法相比，本文模型具有优越性能。值得注意的是，在 KITTI Odometry 数据集上，SCIPaD 在相机位姿估计任务中的平均平移误差降低了 22.2%，平均角度误差降低了 34.8%。源代码可在 https://mias.group/SCIPaD 获取。
英文摘要：
Unsupervised monocular depth estimation frameworks have shown promising performance in autonomous driving. However, existing solutions primarily rely on a simple convolutional neural network for ego-motion recovery, which struggles to estimate precise camera poses in dynamic, complicated real-world scenarios. These inaccurately estimated camera poses can inevitably deteriorate the photometric reconstruction and mislead the depth estimation networks with wrong supervisory signals. In this article, we introduce SCIPaD, a novel approach that incorporates spatial clues for unsupervised depth-pose joint learning. Specifically, a confidence-aware feature flow estimator is proposed to acquire 2D feature positional translations and their associated confidence levels. Meanwhile, we introduce a positional clue aggregator, which integrates pseudo 3D point clouds from DepthNet and 2D feature flows into homogeneous positional representations. Finally, a hierarchical positional embedding injector is proposed to selectively inject spatial clues into semantic features for robust camera pose decoding. Extensive experiments and analyses demonstrate the superior performance of our model compared to other state-of-the-art methods. Remarkably, SCIPaD achieves a reduction of 22.2\% in average translation error and 34.8\% in average angular error for camera pose estimation task on the KITTI Odometry dataset. Our source code is available at \url{https://mias.group/SCIPaD}.
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。
