Monocular image depth estimation method based on pyramid pooling module

Abstract

The invention discloses a monocular image depth estimation method based on a pyramid pooling module. In a training stage, a neural network is firstly constructed, which comprises an input layer, a hidden layer and an output layer. The hidden layer includes a separate first convolution layer, a feature extraction network framework, a scale recovery network framework, a separate second convolution layer, a pyramid pooling module, and a separate connection layer. Each original monocular image in the training set is used as the original input image. The optimal weight vector and the optimal bias term of the trained neural network model are obtained by calculating the loss function value between the predicted depth image and the real depth image corresponding to each original monocular image inthe training set and inputting it into the neural network for training. In the testing phase, the monocular image to be predicted is input into the neural network model, and the predicted depth imageis obtained by using the optimal weight vector and the optimal bias term. The advantages are high prediction accuracy and low computational complexity.

Description

technical field [0001] The present invention relates to an image depth estimation technology, in particular to a monocular image depth estimation method based on a pyramid pooling module. Background technique [0002] Depth estimation is the process of using one or more images to predict a depth map of a scene. Depth information is an important clue to understand the geometric relationship in the scene, and can be applied to various scenarios such as 3D model reconstruction, stereo matching, human pose estimation, etc. Depth information can be obtained from stereo images or motion sequences containing left and right viewpoints, which provide relatively rich information for understanding depth information in space and time, respectively. In contrast, estimating depth from monocular images is harder and more ambiguous because it does not allow matching in space for stereo images or temporally for motion sequences. In 2016, Laina I and others proposed the Fully Convolutional ...

AI Technical Summary

Problems solved by technology

In contrast, estimating depth from monocular images is harder and more ambiguous because it does not allow matching in space for stereo images or temporally for motion sequences

In 2016, Laina I and others proposed the Fully Convolutional Residual Networks (FCRN) framework, which achieved better depth estimation by reducing the size of the image and restoring it. However, due to its relatively simple structure, The advanced features extracted by the neural network are not fully utilized, so there is still room for improvement in prediction accuracy

Images