University of Chinese Academy of Sciences, Beijing 100049, China.
Shenyang Institute of Computing Technology, Chinese Academy of Sciences, Shenyang 110168, China.
Sensors (Basel). 2021 Oct 20;21(21):6956. doi: 10.3390/s21216956.
In recent years, self-supervised monocular depth estimation has gained popularity among researchers because it uses only a single camera at a much lower cost than the direct use of laser sensors to acquire depth. Although monocular self-supervised methods can obtain dense depths, the estimation accuracy needs to be further improved for better applications in scenarios such as autonomous driving and robot perception. In this paper, we innovatively combine soft attention and hard attention with two new ideas to improve self-supervised monocular depth estimation: (1) a soft attention module and (2) a hard attention strategy. We integrate the soft attention module in the model architecture to enhance feature extraction in both spatial and channel dimensions, adding only a small number of parameters. Unlike traditional fusion approaches, we use the hard attention strategy to enhance the fusion of generated multi-scale depth predictions. Further experiments demonstrate that our method can achieve the best self-supervised performance both on the standard KITTI benchmark and the Make3D dataset.
近年来，自监督单目深度估计在研究人员中变得流行起来，因为它仅使用单个相机，成本远低于直接使用激光传感器来获取深度。尽管单目自监督方法可以获得密集的深度，但为了在自动驾驶和机器人感知等场景中更好地应用，估计精度需要进一步提高。在本文中，我们创新性地将软注意和硬注意与两个新思想结合起来，以提高自监督单目深度估计：（1）软注意模块和（2）硬注意策略。我们将软注意模块集成到模型架构中，以增强空间和通道维度的特征提取，仅增加少量参数。与传统的融合方法不同，我们使用硬注意策略来增强生成的多尺度深度预测的融合。进一步的实验表明，我们的方法在标准 KITTI 基准和 Make3D 数据集上都能实现最佳的自监督性能。
