Efficient and High-Quality Monocular Depth Estimation via Gated Multi-Scale Network

• Published in: IEEE access Vol. 8; pp. 7709 - 7718
• Main Authors: Lin, Lixiong, Huang, Guohui, Chen, Yanjie, Zhang, Liwei, He, Bingwei
• Format: Journal Article
• Language: English
• Published: Piscataway IEEE 2020; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Blurring; Convolution; Decoding; Deep learning; Depth estimation; Estimation; Feature extraction; gated multi-scale network; Image manipulation; Image quality; Image reconstruction; Image resolution; Logic gates; Model accuracy; Modules; monocular vision; super resolution

The key issue in monocular depth estimation is how to construct the depth image better and improve the quality of the depth map. At present, most of the monocular depth estimation methods based on deep learning manipulate images at low resolution that leads to loss of detail and blurring of boundari...