Full convolution neural network (FCN)-based monocular image depth estimation method

Abstract

The invention discloses a full convolution neural network (FCN)-based monocular image depth estimation method. The method comprises the steps of acquiring training image data; inputting the training image data into a full convolution neural network (FCN), and sequentially outputting through pooling layers to obtain a characteristic image; subjecting each characteristic image outputted by a last pooling layer sequentially to amplification treatment to obtain a new characteristic image the same with the dimension of a characteristic image outputted by a previous pooling layer, and fusing the twocharacteristic images; sequentially fusing the outputted characteristic image of each pooling layer from back to front so as to obtain a final prediction depth image; training the parameters of the full convolution neural network (FCN) by utilizing a random gradient descent method (SGD) during training; acquiring an RGB image required for depth prediction, and inputting the RGB image into the well trained full convolution neural network (FCN) so as to obtain a corresponding prediction depth image. According to the method, the problem that the resolution of an output image is low in the convolution process can be solved. By adopting the form of the full convolution neural network, a full-connection layer is removed. The number of parameters in the network is effectively reduced.

Description

technical field

[0001] The invention relates to a monocular image depth estimation method based on a fully convolutional neural network (FCN), and belongs to the technical field of three-dimensional image reconstruction of computer vision. Background technique

[0002] Recovering 3D depth information from 2D images is an important problem in the field of computer vision and an essential part of understanding scene geometry. Image depth information has important applications in robotics, scene understanding, 3D reconstruction, etc. The acquisition of image depth information aims to obtain the spatial position information between different objects in the image. Currently, there are two main ways to obtain image depth information. One is to directly obtain depth information through hardware devices, such as Kinect. Another widely used method is to use single or multiple RGB image sequences of the same scene for depth estimation, including multi-view, binocular and single-vie...

Images