A foveated event-camera OCR pipeline reconstructs binary text images and reads them with an LLM, achieving 8.3% word error at 30 lux where an RGB camera fails.
Eventstr: A benchmark dataset and baselines for event stream based scene text recognition, 2025
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
baseline 1
citation-polarity summary
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1roles
baseline 1polarities
baseline 1representative citing papers
citing papers explorer
-
Reading in the Dark with Foveated Event Vision
A foveated event-camera OCR pipeline reconstructs binary text images and reads them with an LLM, achieving 8.3% word error at 30 lux where an RGB camera fails.