pith. sign in

arxiv: 2301.06715 · v1 · pith:NLGXKD4Jnew · submitted 2023-01-17 · 💻 cs.CV · cs.LG· cs.RO

SwinDepth: Unsupervised Depth Estimation using Monocular Sequences via Swin Transformer and Densely Cascaded Network

classification 💻 cs.CV cs.LGcs.RO
keywords depthestimationunsupervisedcascadeddenselymonocularnetworkalgorithms
0
0 comments X
read the original abstract

Monocular depth estimation plays a critical role in various computer vision and robotics applications such as localization, mapping, and 3D object detection. Recently, learning-based algorithms achieve huge success in depth estimation by training models with a large amount of data in a supervised manner. However, it is challenging to acquire dense ground truth depth labels for supervised training, and the unsupervised depth estimation using monocular sequences emerges as a promising alternative. Unfortunately, most studies on unsupervised depth estimation explore loss functions or occlusion masks, and there is little change in model architecture in that ConvNet-based encoder-decoder structure becomes a de-facto standard for depth estimation. In this paper, we employ a convolution-free Swin Transformer as an image feature extractor so that the network can capture both local geometric features and global semantic features for depth estimation. Also, we propose a Densely Cascaded Multi-scale Network (DCMNet) that connects every feature map directly with another from different scales via a top-down cascade pathway. This densely cascaded connectivity reinforces the interconnection between decoding layers and produces high-quality multi-scale depth outputs. The experiments on two different datasets, KITTI and Make3D, demonstrate that our proposed method outperforms existing state-of-the-art unsupervised algorithms.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.