Mohamed N. Sweilam^1,2,* and Nikolay Tolstokulakov², ¹Suez University, Egypt, ²Novosibirsk State University, Russia

Depth estimation has made great progress in the last few years due to its applications in robotics science and computer vision. Various methods have been developed and implemented to estimate the depth, without flickers and missing holes. Despite this progress, it is still one of the main challenges for researchers, especially for the video applications which have more difficulties such as the complexity of the neural network which affects the run time. Moreover to use such input like monocular video for depth estimation is considered an attractive idea, particularly for hand-held devices such as mobile phones, nowadays they are very popular for capturing pictures and videos. Here in this work, we focus on enhancing the existing consistent depth estimation for monocular videos approach to be with less usage of memory and with using less number of parameters without having a significant reduction in the quality of the depth estimation.

Monocular video, monocular depth estimation, deep learning, geometric consistency, lightweight network.