US 2021/0201009 A1 · PEDESTRIAN RE-IDENTIFICATION METHOD AND APPARATUS

Abstract

The present disclosure provides a pedestrian re-identification method and apparatus. The pedestrian re-identification apparatus detects a pedestrian to be identified in a preset video frame, extract all pedestrians appearing in video frames shot in a preset time period before a shooting time of the preset video frame as historical pedestrians, calculate a feature distance between the pedestrian to be identified and each historical pedestrian of the historical pedestrians, extract historical pedestrian IDs corresponding to a preset number of the feature distances in ascending order of the feature distances, and cluster the historical pedestrian IDs to re-identify the pedestrian to be identified.

Claims14

Dependent claims

A pedestrian re-identification method, comprising:

detecting a pedestrian to be identified in a preset video frame;
extracting all pedestrians appearing in video frames shot in a preset time period before a shooting time of the preset video frame as historical pedestrians, wherein an end time of the preset time period is the shooting time of the preset video frame;
calculating a feature distance between the pedestrian to be identified and each historical pedestrian of the historical pedestrians;
extracting historical pedestrian IDs corresponding to a preset number of the feature distances in ascending order of the feature distances;
and
clustering the historical pedestrian IDs to re-identify the pedestrian to be identified.

3 more elements

A pedestrian re-identification apparatus, comprising:

a processor;
and
a memory coupled to the processor, storing program instructions which, when executed by the processor, cause the processor to:
detect a pedestrian to be identified in a preset video frame;
extract all pedestrians appearing in video frames shot in a preset time period before a shooting time of the preset video frame as historical pedestrians, wherein an end time of the preset time period is the shooting time of the preset video frame;
calculate a feature distance between the pedestrian to be identified and each historical pedestrian of the historical pedestrians;
extract historical pedestrian IDs corresponding to a preset number of the feature distances in ascending order of the feature distances;
and
cluster the historical pedestrian IDs to re-identify the pedestrian to be identified.

6 more elements

Dependencies are inferred from the claim text.

Facts

Application

17/044,578Filed 26 Apr 2019

Priority

CN 201810383810.826 Apr 2018

Published

1 Jul 2021

Term ends

Not estimatedThe source records don’t support an estimate.

Classification9 codes

G06V 40/103
Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands: static body considered as a whole, e.g. static pedestrian…
G06F 18/22
Analysing: matching criteria, e.g. proximity measures
G06F 18/23
Analysing: clustering techniques
G06V 10/762
Using pattern recognition or machine learning: using clustering, e.g. of similar faces in social networks

Pedestrian re-identification method and apparatus

Life of this patent

Pedestrian re-identification method and apparatus

Life of this patent

Abstract

Claims14