A binocular depth estimation method based on depth neural network

Abstract

The invention relates to a binocular depth estimation method based on a depth neural network, which comprises the following steps: 1) preprocessing the input left and right viewpoint images to enhancedata; 2) constructing a multi-scale network model for binocular depth estimation, wherein the model comprises a plurality of convolution layers, an activation layer, a residual connection, a multi-scale pooling connection and a linear upsampling layer; 3) Designing the loss function to minimize the results in the continuous training process, so as to obtain the optimal network weights; 4) inputting the image to be processed into the network model to obtain the corresponding depth map, and repeatedly repeating the above steps until the network converges or reaches the training times. The invention adopts the idea of unsupervised learning, and only the left and right viewpoint images obtained by the binocular camera are used as network input. The adaptive design of the network sets the internal and external parameters of the camera as a single model parameter, so it can be applied to multiple camera systems without modifying the network.

Description

technical field [0001] The invention belongs to the field of multimedia image processing, relates to computer vision and deep learning technology, and is a binocular depth estimation method based on a deep neural network. technical background [0002] Depth estimation has always been a popular research direction in the field of computer vision. The 3D data provided by the depth map provides the required information for the realization of 3D reconstruction, augmented reality (Augmented Reality, AR), intelligent navigation and other applications. At the same time, the positional relationship expressed by the depth map is extremely important in many image tasks, which can further simplify the image processing algorithm. At present, the more common depth estimation is mainly divided into two categories, namely, monocular depth estimation and binocular depth estimation. [0003] The monocular depth estimation method only uses one camera. In the traditional algorithm, the camera ...

AI Technical Summary

Problems solved by technology

[0004] However, supervised learning usually relies too much on the real value, but the real value may have errors and noise, the depth information is relatively sparse, and the hardware device is difficult to calibrate, which makes the estimated depth value inaccurate.

Images