Multi-object de-duplication method for real-time video, terminal device and storage medium

A real-time video, multi-target technology, applied in the field of computer vision, can solve the problems of unsuitable application scenarios, high computational complexity, limited accuracy, etc., to reduce redundant work, improve robustness, and ensure real-time effects.
CN109543641AActive Publication Date: 2019-03-29XIAMEN MEIYA PICO INFORMATION
8 Cites 13 Cited by

Patent Information

Authority / Receiving Office
CN · China
Current Assignee / Owner
XIAMEN MEIYA PICO INFORMATION
Publication Date
2019-03-29

Smart Images

  • Figure 1
    Figure 1
  • Figure 2
    Figure 2
  • Figure 3
    Figure 3
Patent Text Reader

Abstract

The invention relates to a multi-object de-duplication method of real-time video, a terminal device and a storage medium. In the method, the multi-object de-duplication method comprises the followingsteps: S100, extracting pictures of each object in the picture according to a video frame picture at a current time; S200: obtaining corresponding structured information data according to pictures ofeach target: a target image feature vector, tracking information and an image quality score; S300: setting a set of structured information data of each target before the current time as a target set,carrying out Clustering according to the structured information data of the targets, and adding the structured information data of each target at the current time to the target set, so that the same target appears only once at different times in the target set, and the appearing target is the target with the best relative quality. The method realizes the output result of the target detection through clustering, and only one picture of the same object is output finally, which greatly reduces the redundancy of the system, reduces the error judgment rate of the detection in a certain range, and improves the robustness.
Need to check novelty before this filing date? Find Prior Art

Description

technical field

[0001] The present invention relates to the technical field of computer vision, in particular to a multi-object deduplication method for real-time video, a terminal device and a storage medium. Background technique

[0002] Object detection is one of the basic tasks in the field of computer vision. Its main job is to identify multiple objects in a picture and locate their positions in the image. The current popular methods for extracting structured information from video mainly rely on the target detection model to mark the target of interest first, and then further analyze the features of each marked target to determine the color, shape and other more specific information of the target. .

[0003] The output of the video frame after target detection is directly used as the input of subsequent applications (such as face recognition, license plate recognition), and multiple repeated pictures of the same target will be collected, which is generally unavoidable...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More