Cross-modal pedestrian re-identification model and method based on auxiliary features and application

The invention provides a cross-modal pedestrian re-identification model and method based on auxiliary features and application, and the method comprises the steps: obtaining the annotation data from an existing cross-modal pedestrian re-identification data set, inputting the annotation data into a c...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: CHAN SIXIAN, SUN JINGJING, QI MENGZAN, WU ZHOUJIAN
Format: Patent
Sprache:chi ; eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:The invention provides a cross-modal pedestrian re-identification model and method based on auxiliary features and application, and the method comprises the steps: obtaining the annotation data from an existing cross-modal pedestrian re-identification data set, inputting the annotation data into a cross-modal pedestrian re-identification frame based on the auxiliary features, and carrying out the training of the cross-modal pedestrian re-identification frame based on the auxiliary features. A self-adaptive feature extraction module is used for deeply mining discriminative features of pedestrians, in order to better relieve modal differences, an auxiliary feature guide module is designed, additional auxiliary modal features are generated, and learning of modal irrelevant features is assisted. In a test stage, the trained network is adopted to extract image features, so that cross-modal image retrieval is realized. According to the invention, efficient extraction of discriminative pedestrian features is realize