当前位置: X-MOL 学术IEEE Signal Process. Lett. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Distortion-Aware Monocular Depth Estimation for Omnidirectional Images
IEEE Signal Processing Letters ( IF 3.9 ) Pub Date : 2021-01-11 , DOI: 10.1109/lsp.2021.3050712
Hong-Xiang Chen , Kunhong Li , Zhiheng Fu , Mengyi Liu , Zonghao Chen , Yulan Guo

Image distortion is a main challenge for tasks on panoramas. In this work, we propose a Distortion-Aware Monocular Omnidirectional (DAMO) network to estimate dense depth maps from indoor panoramas. First, we introduce a distortion-aware module to extract semantic features from omnidirectional images. Specifically, we exploit deformable convolution to adjust its sampling grids to geometric distortions on panoramas. We also utilize a strip pooling module to sample against horizontal distortion introduced by inverse gnomonic projection. Second, we introduce a plug-and-play spherical-aware weight matrix for our loss function to handle the uneven distribution of areas projected from a sphere. Experiments on the 360D dataset show that the proposed method can effectively extract semantic features from distorted panoramas and alleviate the supervision bias caused by distortion. It achieves the state-of-the-art performance on the 360D dataset with high efficiency.

中文翻译:

全方位图像的失真感知单眼深度估计

图像失真是全景任务的主要挑战。在这项工作中,我们提出了一种失真感知单眼全向(DAMO)网络,以从室内全景图中估计密集的深度图。首先,我们引入了一种失真感知模块,以从全向图像中提取语义特征。具体来说,我们利用可变形卷积将其采样网格调整为全景图上的几何变形。我们还利用条带池化模块来采样以防止因反侏儒投影引起的水平失真。其次,我们为损失函数引入了即插即用的球形感知权重矩阵,以处理从球体投影的区域的不均匀分布。在360D数据集上进行的实验表明,该方法可以有效地从扭曲的全景图中提取语义特征,并减轻扭曲引起的监督偏差。它以高效率在360D数据集上实现了最先进的性能。
更新日期:2021-02-16
down
wechat
bug