CPC G06T 7/55 (2017.01) [B60W 60/001 (2020.02); G01S 13/89 (2013.01); G01S 13/931 (2013.01); G06N 3/08 (2013.01); G06T 7/13 (2017.01); B60W 2420/42 (2013.01); B60W 2420/52 (2013.01); G06T 2207/10028 (2013.01); G06T 2207/10044 (2013.01); G06T 2207/20081 (2013.01); G06T 2207/30252 (2013.01)] | 20 Claims |
1. A method for training an image depth estimation model, comprising:
inputting a sample environmental image, sample environmental point cloud data and sample edge information of the sample environmental image into a to-be-trained model; and
determining initial depth information of each of pixel points in the sample environmental and a feature relationship between each of the pixel points and a corresponding neighboring pixel point of each of the pixel points according to the sample environmental image, the sample environmental point cloud data and the sample edge information of the sample environmental image through the to-be-trained model, and optimizing the initial depth information of each of the pixel points according to the feature relationship to obtain optimized depth information of each of the pixel points, and adjusting a parameter of the to-be-trained model according to the optimized depth information to obtain the image depth estimation model;
wherein the to-be-trained model is a neural network model, and the image depth estimation model is configured to obtain first optimized depth information of respective pixel points in an environmental image.
|