Camera-based Kanji OCR for Mobile-phones: Practical Issues
ICDAR '05 Proceedings of the Eighth International Conference on Document Analysis and Recognition
Mosaicing-by-recognition for video-based text recognition
Pattern Recognition
Latency insensitive task scheduling for real-time video processing and streaming
ACIVS'05 Proceedings of the 7th international conference on Advanced Concepts for Intelligent Vision Systems
Predictive directional rectangular zonal search for digital multimedia processor
PCM'05 Proceedings of the 6th Pacific-Rim conference on Advances in Multimedia Information Processing - Volume Part I
Hi-index | 0.00 |
We propose a new technology called the Camera-TypingInterface, which can read printed characterssuch as URLs with a low-resolution camera. It realizesa type of ubiquitous information service using smartphones with cameras. This method includes two mainadvantages: automatic concatenation of sequentialshots and automatic error correction by re-shooting.The automatic concatenation enables a user to takesegmented images of a character string, thus a low-resolutioncamera can be used as the interface device.The automatic error correction enables the user tocorrect misrecognized characters just by retakingimages around them, providing an easy and naturalway of error correction. We present two experimentalresults to prove effectiveness of our method. Bothresults indicate that our proposed method is helpful.