View source for 17525250. SYSTEM AND METHODS FOR MULTIPLE INSTANCE SEGMENTATION AND TRACKING simplified abstract (HUAWEI TECHNOLOGIES CO., LTD.)

=SYSTEM AND METHODS FOR MULTIPLE INSTANCE SEGMENTATION AND TRACKING=


==Organization Name==

[[:Category:HUAWEI TECHNOLOGIES CO., LTD.|HUAWEI TECHNOLOGIES CO., LTD.]]


[[Category:HUAWEI TECHNOLOGIES CO., LTD.]]

==Inventor(s)==

[[:Category:Qiang Tang of Burnaby (CA)|Qiang Tang of Burnaby (CA)]][[Category:Qiang Tang of Burnaby (CA)]]

[[:Category:Vishnu Sanjay Ramiya Srinivasan of Scarborough (CA)|Vishnu Sanjay Ramiya Srinivasan of Scarborough (CA)]][[Category:Vishnu Sanjay Ramiya Srinivasan of Scarborough (CA)]]

[[:Category:Shao Hua Chen of Surrey (CA)|Shao Hua Chen of Surrey (CA)]][[Category:Shao Hua Chen of Surrey (CA)]]

[[:Category:Zhan Xu of Richmond (CA)|Zhan Xu of Richmond (CA)]][[Category:Zhan Xu of Richmond (CA)]]

==SYSTEM AND METHODS FOR MULTIPLE INSTANCE SEGMENTATION AND TRACKING - A simplified explanation of the abstract==


This abstract first appeared for US patent application 17525250 titled 'SYSTEM AND METHODS FOR MULTIPLE INSTANCE SEGMENTATION AND TRACKING

==Simplified Explanation==

The abstract of this patent application describes a method and system for multiple instance segmentation and tracking. Here is a simplified explanation of the abstract:

* The method involves sending an image to a backbone network to generate image feature outputs.
* These image feature outputs are then sent to a spatial attention module, which generates a feature map that identifies objects in the image.
* The feature map is also sent to a category feature module, which generates an instance category output indicating the objects present in the image.
* Additionally, the image feature outputs are sent to a mask generating module, which generates masks for the identified objects.
* The method further includes generating re-identification embedding information associated with the objects based on the image feature outputs.

Potential applications of this technology:

* Object detection and tracking in surveillance systems.
* Autonomous vehicles for identifying and tracking multiple objects on the road.
* Robotics for object recognition and manipulation in complex environments.
* Augmented reality applications for real-time object segmentation and tracking.

Problems solved by this technology:

* Efficiently identifying and tracking multiple objects in an image or video.
* Accurately generating masks for each identified object.
* Providing instance category output to indicate the type or category of each object.
* Generating re-identification embedding information for object tracking purposes.

Benefits of this technology:

* Improved accuracy and efficiency in multiple instance segmentation and tracking.
* Enhanced object recognition and tracking capabilities in various applications.
* Real-time processing of images or videos for immediate object identification and tracking.
* Potential for integration with existing systems and technologies.


==Original Abstract Submitted==

This disclosure provides for methods and a system for multiple instance segmentation and tracking. According to an aspect a method is provided. The method includes sending an image to a backbone network and generating image feature outputs. The method further includes sending the image feature outputs to a spatial attention module for generating a feature map associated with objects in the image. The method further includes sending the feature map to a category feature module for generating an instance category output indicating the objects. The method further includes sending the image feature outputs to a mask generating module for generating masks. The method further includes generating: the instance category output via the category feature module, and the masks via the mask generating module. In some embodiments, the method further includes generating re-identification embedding information associated with the objects based on image feature outputs.



[[Category:G06T5/00]]
[[Category:G06T7/136]]
[[Category:G06K9/62]]
[[Category:G06N3/04]]