Project Details

Description

This projects deals with vision-based person re-identification to measure queue times in, e.g., an airport. The primary goal of this project is to develop very discriminative features to distinguish persons that have been captured by non-overlapping cameras. Having cameras located at the queue entrance and exit, features from the corresponding views can be matched to find the time from a person entered the queue to leaving.
StatusFinished
Effective start/end date01/01/201731/12/2019

Keywords

  • person re-identification
  • multimodal
  • convolution neural network
  • feature fusion

Research Output

  • 2 Article in proceeding
  • 2 Journal article

Person Re-identification Using Spatial and Layer-Wise Attention

Lejbølle, A. R., Nasrollahi, K., Krogh, B. & Moeslund, T. B., 5 Sep 2019, In : I E E E Transactions on Information Forensics and Security. 15, p. 1216 - 1231 16 p., 8826013.

Research output: Contribution to journalJournal articleResearchpeer-review

Open Access
File
  • 40 Downloads (Pure)

    Attention in Multimodal Neural Networks for Person Re-identification

    Lejbølle, A. R., Krogh, B., Nasrollahi, K. & Moeslund, T. B., Jun 2018, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW). IEEE, p. 292-300 9 p. (IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)).

    Research output: Contribution to book/anthology/report/conference proceedingArticle in proceedingResearchpeer-review

    Open Access
    File
  • 2 Citations (Scopus)
    305 Downloads (Pure)

    Enhancing Person Re-identification by Late Fusion of Low-, Mid-, and High-Level Features

    Lejbølle, A. R., Nasrollahi, K. & Moeslund, T. B., 2018, In : IET Biometrics. 7, 2, p. 125-135 22 p.

    Research output: Contribution to journalJournal article

    Open Access
    File
  • 221 Downloads (Pure)