Unlock instant, AI-driven research and patent intelligence for your innovation.

Image frame depth estimation method and device, electronic equipment and storage medium

A technology of depth estimation and image frame, applied in the computer field, it can solve the problems that PoseNet cannot give and the generalization performance is limited, so as to improve the robustness, application performance and generalization ability.

Active Publication Date: 2020-06-26
TSINGHUA UNIV +1
View PDF3 Cites 18 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, using PoseNet to predict camera pose changes is not robust enough. On the distribution of pose data that has not appeared in the training set, the trained PoseNet cannot give effective predictions, indicating that its generalization performance is limited.

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Image frame depth estimation method and device, electronic equipment and storage medium
  • Image frame depth estimation method and device, electronic equipment and storage medium
  • Image frame depth estimation method and device, electronic equipment and storage medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0050] The specific embodiments of the present invention will be further described below in conjunction with the accompanying drawings. The following examples are only used to illustrate the technical solution of the present invention more clearly, but not to limit the protection scope of the present invention.

[0051] figure 1 It shows a schematic flowchart of a method for estimating the depth of an image frame provided by this embodiment, including:

[0052] S101. Obtain two adjacent image frames from the training video sequence, and respectively input the two image frames into the optical flow prediction network obtained through unsupervised training, and obtain the two images output by the optical flow prediction network Correspondence between all pixels of the frame.

[0053] Wherein, the training video sequence is a video sequence used for training the optical flow prediction network and the depth prediction network.

[0054] The optical flow prediction network and t...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The embodiment of the invention discloses an image frame depth estimation method and device, electronic equipment and a storage medium. An optical flow prediction network obtained through unsupervisedtraining is used for extracting a pixel relationship to replace traditional manually set image features SIFT and the like for matching, the determination of the relationship between pixels becomes more accurate, and meanwhile confidence sampling is introduced to further improve robustness; and the camera pose relationship is solved by establishing the pixel relationship, and end-to-end camera pose relative change estimation is replaced, so that the generalization ability is greatly improved, and the application performance of the whole system is improved.

Description

technical field [0001] The present invention relates to the field of computer technology, in particular to a method, device, electronic equipment and storage medium for estimating the depth of an image frame. Background technique [0002] Depth estimation of monocular images is a hot issue in computer vision and robotics, and has a wide range of applications in areas such as autonomous driving, robot navigation, and 3D reconstruction; visual odometry, or camera pose estimation, is an important hotspot in the field of robotics question. In video sequences, depth information and camera pose information constrain and influence each other, so the combined solution and application of depth estimation and pose estimation has attracted more and more attention. [0003] The development of neural networks and the emergence of deep learning have brought new ideas and solutions to traditional computer vision tasks. It becomes possible to learn to estimate depth information from large...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G06T7/55
CPCG06T7/55G06T2207/10016G06T2207/20081G06T2207/20084
Inventor 刘永进赵旺舒叶芷
Owner TSINGHUA UNIV