Reproduce “Luggage re-identification: A study of strategies towards a real-world application” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Quentin Dhôte, Romaric Audigier
The authors present an airport luggage retrieval system that matches corridor surveillance tracks to conveyor-belt enrollment views. Because bags are hidden and tracks fragmented, the system keeps corridor tracks camera-specific, links gallery views across cameras, and averages five confidence-selected query observations. On their real-airport dataset, Q-to-G R@1 reaches 0.812, versus 0.794 for a single-frame baseline; this does not establish generalization without adaptation.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.