Vision Geometry XIII

TL;DR: This paper summarizes the recent efforts on offline handwritten Chinese script recognition using a segmentation-driven approach and proposes a heteroscedastic linear discriminant analysis algorithm to extract more discrimination information from original character features, and implement a minimum classification error learning scheme to optimize classifier parameters.

TL;DR: In this article, the authors tested whether the use of known contemporaneous mechanical methods might explain both image evidence and optical methods, and found that the fidelities of these modern re-enactments were equal or superior to those of the van Eyck works.