Characters Detection on Namecard with faster RCNN
This work addresses the problem of character detection for namecard digitization, but it is incremental as it applies standard methods to a specific domain with minor enhancements.
The paper tackled character detection on namecards using Faster R-CNN, achieving an mAP of 80% and average IoU of at least 80% without data augmentation, with slight improvements in both metrics after applying data augmentation techniques.
We apply Faster R-CNN to the detection of characters in namecard, in order to solve the problem of a small amount of data and the inbalance between different class, we designed the data augmentation and the 'fake' data generalizer to generate more data for the training of network. Without using data augmentation, the average IoU in correct samples could be no less than 80% and the mAP result of 80% was also achieved with Faster R-CNN. By applying the data augmentation, the variance of mAP is decreased and both of the IoU and mAP score has increased a little.