URCDC-深度：基于CutFlip的不确定性校正交叉蒸馏用于单目深度估计
URCDC-Depth: Uncertainty Rectified Cross-Distillation with CutFlip for Monocular Depth Estimation
摘要：
这项工作旨在从单个RGB图像中估计高质量的深度图。由于缺乏深度线索，充分利用长程相关和局部信息是精确深度估计的关键。为此，我们在Transformer和卷积神经网络（CNN）之间引入了一种不确定性校正交叉蒸馏，以学习统一的深度估计器。具体来说，我们使用来自Transformer分支和CNN分支的深度估计作为伪标签来相互教导。同时，我们对像素级深度不确定性进行建模，以纠正噪声伪标签的损失权重。为了避免由强Transformer分支引起的大容量间隙恶化交叉蒸馏，我们将特征图从Transformer转移到CNN，并设计耦合单元来帮助弱CNN分支利用转移的特征。此外，我们提出了一个令人惊讶的简单而高效的数据增强技术CutFlip，它强制执行的模型，以利用更有价值的线索，除了垂直图像位置的深度推断。大量的实验表明，我们的模型，称为~ \textbf{URCDC-Depth}，超过了以前的国家的最先进的方法在KITTI，NYU-Depth-v2和SUN RGB-D数据集，即使没有额外的计算负担在推理时间。源代码可在\url{https://github.com/ShuweiShao/URCDC-Depth}上公开获得。
英文摘要：
This work aims to estimate a high-quality depth map from a single RGB image. Due to the lack of depth clues, making full use of the long-range correlation and the local information is critical for accurate depth estimation. Towards this end, we introduce an uncertainty rectified cross-distillation between Transformer and convolutional neural network (CNN) to learn a unified depth estimator. Specifically, we use the depth estimates from the Transformer branch and the CNN branch as pseudo labels to teach each other. Meanwhile, we model the pixel-wise depth uncertainty to rectify the loss weights of noisy pseudo labels. To avoid the large capacity gap induced by the strong Transformer branch deteriorating the cross-distillation, we transfer the feature maps from Transformer to CNN and design coupling units to assist the weak CNN branch to leverage the transferred features. Furthermore, we propose a surprisingly simple yet highly effective data augmentation technique CutFlip, which enforces the model to exploit more valuable clues apart from the vertical image position for depth inference. Extensive experiments demonstrate that our model, termed~\textbf{URCDC-Depth}, exceeds previous state-of-the-art methods on the KITTI, NYU-Depth-v2 and SUN RGB-D datasets, even with no additional computational burden at inference time. The source code is publicly available at \url{https://github.com/ShuweiShao/URCDC-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。
