English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Most current multi-object trackers focus on short-term tracking, and arebased on deep and complex systems that do not operate in real-time, oftenmaking them impractical for video-surveillance. In this paper, we present along-term multi-face tracking architecture conceived for working in crowdedcontexts, particularly unconstrained in terms of movement and occlusions, andwhere the face is often the only visible part of the person. Our systembenefits from advances in the fields of face detection and face recognition toachieve long-term tracking. It follows a tracking-by-detection approach,combining a fast short-term visual tracker with a novel online trackletreconnection strategy grounded on face verification. Additionally, a correctionmodule is included to correct past track assignments with no extracomputational cost. We present a series of experiments introducing novel,specialized metrics for the evaluation of long-term tracking capabilities and avideo dataset that we publicly release. Findings demonstrate that, in thiscontext, our approach allows to obtain up to 50% longer tracks thanstate-of-the-art deep learning trackers.

Long-Term Face Tracking for Crowded Video-Surveillance Scenarios