Offline Writer Identification based on the Path Signature Feature
This work addresses the problem of identifying writers from scanned handwriting for applications in forensics and document analysis, but it is incremental as it builds on existing path signature methods with a codebook approach.
The paper tackled offline writer identification by proposing a novel set of features based on the path signature approach, which extracts local pathlets from handwriting contours to characterize writing style, and achieved competitive results on benchmark datasets including IAM, Firemaker, CVL, and ICDAR2013.
In this paper, we propose a novel set of features for offline writer identification based on the path signature approach, which provides a principled way to express information contained in a path. By extracting local pathlets from handwriting contours, the path signature can also characterize the offline handwriting style. A codebook method based on the log path signature---a more compact way to express the path signature---is used in this work and shows competitive results on several benchmark offline writer identification datasets, namely the IAM, Firemaker, CVL and ICDAR2013 writer identification contest dataset.