A Comparative Study of Face Re-identification Systems under Real-World Conditions

Research output: Chapter in Book/Report/Conference proceedingConference contribution

43 Downloads (Pure)

Abstract

Face re-identification is largely thought of as a solved problem in the research community, with State-of-the-art systems attaining human-level performance on unconstrained image datasets. However, these results do not seem to translate to the real world. In systems matching people in surveillance-like footage to high-quality images, reported performance is much lower than what the literature would suggest. In this work, we contribute a multi-modal dataset for evaluating real-world performance of facial re-identification systems. We then perform a verification and re-identification evaluation for state-of-art systems on both this dataset and the popular benchmarking dataset Labelled Faces in the Wild (LFW).
Original languageEnglish
Title of host publicationIrish Machine Vision and Image Processing Conference Proceedings 2018: Proceedings
PublisherIrish Pattern Recognition & Classification Society
Pages137-144
ISBN (Electronic)978-0-9934207-3-3
Publication statusPublished - 31 Aug 2018
EventIrish Machine Vision and Image Processing Conference - Ulster University, Belfast, United Kingdom
Duration: 29 Aug 201831 Aug 2018
Conference number: 2018

Conference

ConferenceIrish Machine Vision and Image Processing Conference
Abbreviated titleIMVIP
CountryUnited Kingdom
CityBelfast
Period29/08/201831/08/2018

Fingerprint Dive into the research topics of 'A Comparative Study of Face Re-identification Systems under Real-World Conditions'. Together they form a unique fingerprint.

  • Cite this

    Brown, G., Martinez del Rincon, J., & Miller, P. (2018). A Comparative Study of Face Re-identification Systems under Real-World Conditions. In Irish Machine Vision and Image Processing Conference Proceedings 2018: Proceedings (pp. 137-144). Irish Pattern Recognition & Classification Society. https://iprcs.scss.tcd.ie/pdf/IMVIP2018Book.pdf#page=147