What question did this study set out to answer?

The aim is to develop a real-time OCR pipeline that ensures predictable timing for edge vision applications.

April 25, 2026Open Access

RT-OCR : Real-Time OCR Pipeline with Formal Timing Guarantees for Edge Vision Systems

Key Points

The aim is to develop a real-time OCR pipeline that ensures predictable timing for edge vision applications.
Developed a timing-aware pipeline architecture for OCR on edge devices.
Implemented an adaptive workload controller to manage processing demands.
Conducted performance evaluation on Jetson Nano and Raspberry Pi 5 with 10,000 consecutive frames.
Achieved 73% reduction in latency jitter.
Improved worst-case execution time (WCET) by 32%.
Demonstrated 97.5% accuracy with 100% compliance to deadlines.

Abstract

Deep learning OCR achieves high accuracy but exhibits unpredictable latency, preventing deployment in safety-critical systems like autonomous vehicles and industrial robotics. RT-OCR provides formal worst-case execution time (WCET) bounds on edge platforms through a timing-aware pipeline architecture, adaptive workload controller, and measurement-based WCET characterization on Jetson Nano and Raspberry Pi 5. Evalua- tion demonstrates 73% jitter reduction, 32% WCET improvement, and 97.5% accuracy with 100% deadline compliance across 10,000 consecutive frames.

Read Full Paperexternally

Connected Papers

Building similarity graph...

Analyzing shared references across papers

Discussion

References and Citations

Add This Paper to Your Research Feed

Any time a new paper drops it will be there.