Trained correlated-photon illumination paired with a Transformer backend improves object classification accuracy by up to 15 percentage points in photon-starved noisy imaging.
Tables A1→A4 show the probabilities of measuring|00⟩ → |11⟩from the possible illuminations and object classes
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
-
Ultra-low-light computer vision using trained photon correlations
Trained correlated-photon illumination paired with a Transformer backend improves object classification accuracy by up to 15 percentage points in photon-starved noisy imaging.