论文标题
可变形的空间传播网络，用于深度完成
Deformable spatial propagation network for depth completion
论文作者
论文摘要
由于自动驾驶的发展，最近深度完成引起了广泛的关注，该自动驾驶的发展旨在从稀疏深度测量值中恢复密集的深度图。卷积空间传播网络（CSPN）是此任务中最新的方法之一，该方法采用线性传播模型来优化使用局部环境的粗糙深度图。但是，每个像素的传播发生在固定的接受场中。这可能不是完善的最佳选择，因为不同的像素需要不同的本地上下文。为了解决这个问题，在本文中，我们提出了一个可变形的空间传播网络（DSPN），以适应每个像素的接收场和亲和力矩阵。它允许网络获得更少但相关的像素以进行传播的信息。 Kitti深度完成基准测试的实验结果表明，我们提出的方法实现了最先进的性能。
Depth completion has attracted extensive attention recently due to the development of autonomous driving, which aims to recover dense depth map from sparse depth measurements. Convolutional spatial propagation network (CSPN) is one of the state-of-the-art methods in this task, which adopt a linear propagation model to refine coarse depth maps with local context. However, the propagation of each pixel occurs in a fixed receptive field. This may not be the optimal for refinement since different pixel needs different local context. To tackle this issue, in this paper, we propose a deformable spatial propagation network (DSPN) to adaptively generates different receptive field and affinity matrix for each pixel. It allows the network obtain information with much fewer but more relevant pixels for propagation. Experimental results on KITTI depth completion benchmark demonstrate that our proposed method achieves the state-of-the-art performance.
