View source for 18055722. MATCHING BETWEEN 2D AND 3D FOR DIRECT LOCALIZATION simplified abstract (Microsoft Technology Licensing, LLC)

=MATCHING BETWEEN 2D AND 3D FOR DIRECT LOCALIZATION=


==Organization Name==

[[:Category:Microsoft Technology Licensing, LLC|Microsoft Technology Licensing, LLC]]


[[Category:Microsoft Technology Licensing, LLC]]

==Inventor(s)==

[[:Category:Johannes Lutz Sch�nberger of Zurich (CH)|Johannes Lutz Sch�nberger of Zurich (CH)]][[Category:Johannes Lutz Sch�nberger of Zurich (CH)]]

[[:Category:Rui Wang of Zurich (CH)|Rui Wang of Zurich (CH)]][[Category:Rui Wang of Zurich (CH)]]

[[:Category:Prune Solange Garance Truong of Zurich (CH)|Prune Solange Garance Truong of Zurich (CH)]][[Category:Prune Solange Garance Truong of Zurich (CH)]]

[[:Category:Marc André Léon Pollefeys of Zurich (CH)|Marc André Léon Pollefeys of Zurich (CH)]][[Category:Marc André Léon Pollefeys of Zurich (CH)]]

==MATCHING BETWEEN 2D AND 3D FOR DIRECT LOCALIZATION - A simplified explanation of the abstract==


This abstract first appeared for US patent application 18055722 titled 'MATCHING BETWEEN 2D AND 3D FOR DIRECT LOCALIZATION

==Simplified Explanation==

Determining a location of an entity involves receiving a query with a 2D image of the entity's environment and searching for a match in a 3D map of the environment, which includes a 3D point cloud. The match indicates the location of the entity in the environment. The process includes extracting descriptors from the 2D image (image descriptors) and the 3D point cloud (point cloud descriptors), correlating them to produce correspondences, and estimating the location of the entity using these correspondences.

* Extract descriptors from 2D image and 3D point cloud.
* Correlate image descriptors with point cloud descriptors to find correspondences.
* Estimate entity's location using the correspondences.

== Potential Applications ==
This technology could be applied in:
* Augmented reality navigation systems.
* Indoor positioning systems for large buildings.

== Problems Solved ==
This technology helps in:
* Precisely locating entities in complex environments.
* Improving accuracy of location-based services.

== Benefits ==
The benefits of this technology include:
* Enhanced user experience in navigation applications.
* Efficient tracking of assets in industrial settings.

== Potential Commercial Applications ==
A potential commercial application for this technology could be:
* Integration into mobile mapping apps for accurate location tracking.

== Possible Prior Art ==
One possible prior art for this technology could be:
* Similar methods used in robotics for mapping and localization.

== Unanswered Questions ==

=== How does this technology handle occlusions in the environment? ===
The technology may struggle with occlusions that block the view of certain parts of the environment, potentially affecting the accuracy of the entity's location estimation.

=== What is the computational complexity of the matching process? ===
The computational resources required for extracting descriptors, correlating them, and estimating the entity's location may vary depending on the size and complexity of the environment, raising questions about the scalability of the technology.


==Original Abstract Submitted==

Determining a location of an entity comprises: receiving a query comprising a 2D image depicting an environment of the entity; searching for a match between the query and a 3D map of the environment. The 3D map comprising a 3D point cloud, the match indicating the location of the entity in the environment. Searching for the match comprises: extracting descriptors from the 2D image referred to as image descriptors; extracting descriptors from the 3D point cloud referred to as point cloud descriptors; correlating the image descriptors with the point cloud descriptors to produce correspondences, wherein a correspondence is an image descriptor corresponding to a point cloud descriptor; estimating, using the correspondences, the location of the entity.



[[Category:G06T7/73]]
[[Category:G06F16/535]]
[[Category:G06V20/64]]