View source for 17850545. METHOD, DEVICE, AND COMPUTER PROGRAM PRODUCT FOR TRAINING IMAGE CLASSIFICATION MODEL simplified abstract (Dell Products L.P.)

=METHOD, DEVICE, AND COMPUTER PROGRAM PRODUCT FOR TRAINING IMAGE CLASSIFICATION MODEL=


==Organization Name==

[[:Category:Dell Products L.P.|Dell Products L.P.]]


[[Category:Dell Products L.P.]]

==Inventor(s)==

[[:Category:Bin He of Shanghai (CN)|Bin He of Shanghai (CN)]][[Category:Bin He of Shanghai (CN)]]

[[:Category:Zijia Wang of WeiFang (CN)|Zijia Wang of WeiFang (CN)]][[Category:Zijia Wang of WeiFang (CN)]]

[[:Category:Zhen Jia of Shanghai (CN)|Zhen Jia of Shanghai (CN)]][[Category:Zhen Jia of Shanghai (CN)]]

[[:Category:Jinpeng Liu of Shanghai (CN)|Jinpeng Liu of Shanghai (CN)]][[Category:Jinpeng Liu of Shanghai (CN)]]

==METHOD, DEVICE, AND COMPUTER PROGRAM PRODUCT FOR TRAINING IMAGE CLASSIFICATION MODEL - A simplified explanation of the abstract==


This abstract first appeared for US patent application 17850545 titled 'METHOD, DEVICE, AND COMPUTER PROGRAM PRODUCT FOR TRAINING IMAGE CLASSIFICATION MODEL

==Simplified Explanation==

The abstract describes a method, device, and computer program for training an image classification model. Here are the key points:

* The method involves training an image classification model to classify pixel points of a sample image into different object classes.
* In each training iteration, the model determines a classification result for the pixel points.
* The first classification result is obtained in the first training iteration, and the second classification result is obtained in the second training iteration.
* Based on these classification results, the method determines recall rates for each object class in both iterations.
* The recall rates represent the model's ability to correctly identify instances of each object class.
* The image classification model is then adjusted based on these recall rates to improve its performance.
* The result is a trained image classification model that can accurately classify objects in images.

Potential applications of this technology:

* Image recognition and classification systems
* Object detection in images and videos
* Autonomous vehicles for identifying and tracking objects on the road
* Surveillance systems for detecting specific objects or individuals

Problems solved by this technology:

* Improves the accuracy and performance of image classification models
* Enables more precise identification and classification of objects in images
* Reduces false positives and false negatives in object detection systems

Benefits of this technology:

* Enhanced accuracy in identifying and classifying objects in images
* Improved performance of image classification models
* Increased reliability and efficiency in object detection systems
* Enables the development of more advanced and reliable computer vision applications.


==Original Abstract Submitted==

Embodiments disclosed herein relate to a method, a device, and a computer program product for training an image classification model. The method includes: determining a first classification result obtained by the image classification model on pixel points of a sample image in a first training iteration, wherein the first classification result indicates that each of the pixel points belongs to one of a plurality of object classes; determining a second classification result obtained by the image classification model on the pixel points of the sample image in a second training iteration; determining, based on the first classification result and the second classification result, a first set of recall rates and a second set of recall rates for the plurality of object classes; and adjusting, based on the first set of recall rates and the second set of recall rates, the image classification model to obtain a trained image classification model.



[[Category:G06V10/764]]
[[Category:G06V10/82]]
[[Category:G06N3/08]]