Gallery-Distance Profiles for Open-Set Person Re-Identification with Frozen Encoders in Camera Networks
Jin Wang, Jiannong Cao, Changlin Song · 2026
Read Gallery-Distance Profiles for Open-Set Person Re-Identification with Frozen Encoders in Camera Networks by Jin Wang, Jiannong Cao, Changlin Song (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-13568453.txt.
Description
Wang, Cao, and Song present a study from twenty twenty-six on spotting unknown people in camera networks. Systems must first decide whether a queried person is in the gallery at all. Nearest-match distance alone gives weak evidence for that call. Their evaluator reads the whole distance profile instead. Nine plain statistics summarize each profile's shape. Practice episodes teach the evaluator what presence looks like. Calibration uses the real gallery without touching the encoder.
Three standard datasets test the approach. Unknown detection beats six rival scorers overall. Gains grow when cameras differ between query and gallery. Thin identity evidence still caps accuracy, they admit.
Catalog record
| CREATOR | Jin Wang, Jiannong Cao, Changlin Song |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 55 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |