Unlock instant, AI-driven research and patent intelligence for your innovation.

Method and device for pedestrian retrieval

A pedestrian and algorithm technology, applied in the field of intelligent video analysis, can solve the problems of inability to support joint retrieval of multiple high-dimensional features and low retrieval accuracy

Active Publication Date: 2020-04-28
HANGZHOU HUAWEI ENTERPRISE COMM TECH
View PDF7 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Therefore, the existing pedestrian retrieval devices can only perform rough retrieval based on sample images, cannot support joint retrieval of multiple high-dimensional features, and the retrieval accuracy is low

Method used

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
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Method and device for pedestrian retrieval
  • Method and device for pedestrian retrieval
  • Method and device for pedestrian retrieval

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0037] An embodiment of the present invention provides a computer device (or system) 100, such as figure 1 shown in figure 1 The computer device (or system) shown is used to implement the pedestrian retrieval device, including at least one processor 101 , a communication bus 102 , a memory 103 and at least one communication interface 104 .

[0038] The processor 101 may be one processor, or may be a general term for multiple processing elements. For example, the processor 101 may be a general-purpose central processing unit (English full name: Central Processing Unit, English abbreviation: CPU), or an application-specific integrated circuit (English full name: application-specific integrated circuit, English abbreviation: ASIC), or One or more integrated circuits for controlling program execution of the present invention, for example: one or more microprocessors (English full name: digital signal processor, English abbreviation: DSP), or, one or more field programmable gate a...

Embodiment 2

[0053] An embodiment of the present invention provides a method for pedestrian retrieval, which is applied to a pedestrian retrieval device, such as figure 2 As shown, the method includes:

[0054] Step 201. Obtain a pedestrian image sequence set from the video frame to be analyzed.

[0055] The pedestrian retrieval device obtains the pedestrian image sequence from the video frame to be analyzed according to the existing pedestrian detection and tracking algorithm, and how to obtain the pedestrian image sequence can refer to the specific method in the prior art, and the present invention will not repeat it here. A pedestrian image sequence set is formed by a plurality of pedestrian image sequences of pedestrians, the pedestrian image sequence represents a plurality of pedestrian images of a pedestrian, and the pedestrian image includes the video frame image of the pedestrian corresponding to the detection frame output by the pedestrian detection and tracking algorithm and A ...

Embodiment 3

[0117] An embodiment of the present invention provides a pedestrian retrieval device 30, such as Figure 6 shown, including:

[0118] A pedestrian image sequence acquisition module 301, configured to acquire a pedestrian image sequence set from the video frame to be analyzed, the pedestrian image sequence set includes a plurality of pedestrian image sequences of pedestrians, and the pedestrian image sequence represents multiple pedestrian images of a pedestrian , the pedestrian image includes a video frame image of the pedestrian and a mask template used to represent the position of the pedestrian;

[0119] The multi-instance feature extraction module 302 is configured to obtain a multi-instance pedestrian image according to a multi-instance feature extraction algorithm for each pedestrian image sequence, and the multi-instance pedestrian image is acquired according to the local block features of the pedestrian images included in the pedestrian image sequence A collection of ...

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

PUM

No PUM Login to View More

Abstract

Embodiments of the present invention provide a method and device for pedestrian retrieval, which relate to the field of intelligent video analysis, and can extract various features of pedestrian images, establish high-dimensional indexes based on various features of pedestrians, and perform joint retrieval based on various features of pedestrians. Effectively improve retrieval accuracy. Including: obtaining a set of pedestrian image sequences from the video frames to be analyzed; for each pedestrian image sequence, obtaining multi-instance pedestrian images according to the multi-instance feature extraction algorithm; obtaining the low-level features of the pedestrians according to the multi-instance pedestrian images of each pedestrian and semantic features; construct a high-dimensional index according to the low-level features and semantic features of each pedestrian; retrieve the high-dimensional index according to at least one of the low-level features of the sample image of the first pedestrian and the semantic features described in the text, and obtain the retrieval result .

Description

technical field [0001] The invention relates to the field of intelligent video analysis, in particular to a pedestrian retrieval method and device. Background technique [0002] Intelligent video analysis technology is a computer that automatically analyzes the video content according to the retrieval conditions, and retrieves images or video clips that meet the retrieval conditions from a large number of videos. Pedestrian retrieval is the "image search" in the field of intelligent video analysis. Specifically, it is to retrieve all images or video clips containing the pedestrian from a large number of images or videos based on all or part of the images of the pedestrian. [0003] Usually, a pedestrian retrieval device includes a feature extraction module and a high-dimensional index module. The feature extraction module is used to extract pedestrian features. The high-dimensional index module is the key to the real-time retrieval of the pedestrian retrieval device, and i...

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

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Patents(China)
IPC IPC(8): G06K9/00
CPCG06V20/48G06V20/46
Inventor 张伟陈茂林骆立俊尤乾坤
Owner HANGZHOU HUAWEI ENTERPRISE COMM TECH