Patent · US Active

Pedestrian re-identification method based on virtual samples

US11837007B2 · kind B2 · utility

0Cited by
0References
7Claims
0Family size

Assignee

Inventors

Key dates

Filing dateJun 20, 2023
Grant dateDec 5, 2023
Priority date
Expiry dateJun 20, 2043

Classification

  • Technology area (CPC Y)Emerging Cross-Sectional Technologies
  • CPC primaryY02T10/40
  • WIPO fieldComputer technology
  • WIPO sectorElectrical engineering

Abstract

This invention proposes a pedestrian re-identification method based on virtual samples, comprising following steps: s1) obtaining virtual persons generated by game engine, and generating the virtual samples with person labels by fusing a background of a target dataset and a pose of real persons through a multi-factor variational generation network; s2) rendering the generated virtual samples according to lighting conditions; s3) sampling the rendered virtual samples according to person attributes of target dataset; s4) constructing a training dataset according to virtual samples obtained by sampling to train a pedestrian re-identification model, and verifying identification effect of the trained model. The present invention uses a virtual image generation framework that integrates translation-rendering-sampling to narrow the distribution between virtual images and real images as much as possible to generate virtual samples, and conduct person re-identification model training, which can be effectively and effectively applied to pedestrian datasets in real scenes.

Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.