Image depth estimation method based on convolutional neural network

Abstract

The invention discloses an image depth estimation method based on a convolutional neural network. The method comprises the following steps: constructing a convolution-deconvolution pair neural networkmodel, wherein the convolution-deconvolution pair neural network model comprises a plurality of different convolutional layers, a plurality of convolution-deconvolution layer pairs and an activationlayer; selecting a training set, and setting training parameters of the convolution-deconvolution pair neural network model; according to the convolution-deconvolution pair neural network model and the training parameters thereof, training the convolution-deconvolution pair neural network model with minimizing of a loss function being a target to form an image depth estimation neural network model; and inputting an image to be processed to the image depth estimation neural network model and outputting a corresponding depth image. The grayscale value of the depth image obtained through the image depth estimation method based on the convolution-deconvolution pair neural network is more accurate, and the sense of depth of the depth image is higher.

Description

technical field [0001] The invention relates to the fields of computer vision and image processing, in particular to an image depth estimation method based on convolution-deconvolution neural network. Background technique [0002] The depth estimation method is used to estimate the depth information of each pixel in the image to be processed, and obtain the global depth map of the image to be processed, which plays an important role in the application fields of computer vision and computer graphics. Current depth estimation methods can be divided into monocular and binocular based on the number of cameras. [0003] Binocular stereo vision uses two cameras to observe the same object from two viewpoints, obtains perception images under different perspectives of the object, and converts the disparity information of matching points into depth by triangulation. The general binocular vision method uses the epipolar geometry to transform the problem into the Euclidean geometry con...

AI Technical Summary

Problems solved by technology

Such an estimation method usually estimates inaccurate depth values, and the depth map layering is not strong.

At the same time, the influence of the edge of the object in the image is not considered, and the edge of the obtained depth map is blurred

Images