SymbolSight optimizes symbol-to-letter mappings via simulated prosthetic vision and bigram statistics, cutting predicted confusion by a median factor of 22 across Arabic, Bulgarian, and English.
Deep learning– based scene simplification for bionic vision
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
SymbolSight: Minimizing Inter-Symbol Interference for Reading with Prosthetic Vision
SymbolSight optimizes symbol-to-letter mappings via simulated prosthetic vision and bigram statistics, cutting predicted confusion by a median factor of 22 across Arabic, Bulgarian, and English.