KappaPlace learns hyperspherical uncertainty for VPR via prototype-anchored supervision on vMF distributions, cutting expected calibration error by up to 50% while preserving retrieval performance.
Gsv-cities: Toward appropriate supervised visual place recognition.Neurocomputing, 2022
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
KappaPlace: Learning Hyperspherical Uncertainty for Visual Place Recognition via Prototype-Anchored Supervision
KappaPlace learns hyperspherical uncertainty for VPR via prototype-anchored supervision on vMF distributions, cutting expected calibration error by up to 50% while preserving retrieval performance.