Viewpoint optimization method and device for three-dimensional reconstruction, equipment and medium

By constructing the cost function between the detection point and the viewpoint and optimizing the viewpoint selection, the problem of low efficiency and poor effect of three-dimensional reconstruction viewpoint evaluation in the dynamic environment in the prior art is solved, efficient and accurate viewpoint generation is achieved, and the overall performance of three-dimensional reconstruction is improved.

CN119722949AActive Publication Date: 2025-03-28GUANGZHOU SAITE INTELLIGENCE TECH CO LTD
3 Cites 1 Cited by

Patent Information

Application Number
CN202411853146.0
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-12-16
Publication Date
2025-03-28
Estimated Expiration
2044-12-16

AI Technical Summary

Technical Problem

The existing three-dimensional reconstruction viewpoint evaluation method has low processing efficiency in dynamic environments and poor viewpoint generation effect.

Method used

By obtaining the point cloud data of the target area, converting it into raster data, and sampling and clustering the viewpoints around the detection point, the cost function between the detection point and the viewpoint is constructed, and the cost-optimized viewpoint is obtained in turn, and added to the total viewpoint set.

Benefits of technology

It realizes real-time generation of efficient viewpoints, improves the accuracy of three-dimensional reconstruction, and is suitable for exploration and reconstruction of autonomous systems in complex environments, improving task execution efficiency and data quality.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The invention discloses a viewpoint optimization method and device for three-dimensional reconstruction, equipment and a medium, and the method comprises the steps: obtaining point cloud data in a target region, and converting the point cloud data into raster data; carrying out viewpoint sampling on each detection point in the raster data, and carrying out clustering processing on sampled viewpoints to obtain a viewpoint clustering set corresponding to each detection point; constructing a cost function between a detection point and a viewpoint; and according to the cost function, sequentially obtaining the viewpoint with the optimal cost in each viewpoint cluster set, and adding the viewpoint with the optimal cost into the total viewpoint set. The method has the advantages that the detection range of the sensor is comprehensively considered, the cost function mode is combined, the viewpoint possibly generating the high-quality point cloud is preferentially selected, and therefore the optimal viewpoint can be generated in real time, the robot is promoted to efficiently conduct exploration and reconstruction in the complex three-dimensional environment, and the real-time performance of the robot is improved. And thus, the overall task execution efficiency and the data quality are improved.
Need to check novelty before this filing date? Find Prior Art