Object tracking method and apparatus
CN115131397BActive Publication Date: 2026-06-05SAMSUNG ELECTRONICS CO LTD
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- SAMSUNG ELECTRONICS CO LTD
- Filing Date
- 2021-08-25
- Publication Date
- 2026-06-05
AI Technical Summary
Technical Problem
Existing neural network models struggle to effectively handle irregularly shaped targets, especially elongated ones, leading to a decline in tracking performance.
Method used
By setting suppression regions to balance target object information and background information in the template image, the template feature map is refined. Deep neural networks are used for feature extraction and similarity calculation, and the template is updated to adapt to the shape changes of the target object, thereby improving tracking accuracy.
Benefits of technology
It improves the accuracy and stability of tracking irregularly shaped objects, reduces performance degradation caused by shape changes, and enhances the robustness of object tracking.
✦ Generated by Eureka AI based on patent content.
Smart Images
Figure CN115131397B_ABST
Abstract
An object tracking method and apparatus are disclosed. A processor-implemented object tracking method includes setting a suppression region in a template image based on a shape of a target box of the template image, refining a template feature map of the template image by suppressing an influence of feature data of the template feature map corresponding to the suppression region, and tracking an object by determining a bounding box corresponding to the target box in a search image based on the refined template feature map.
Need to check novelty before this filing date? Find Prior Art