This paper propose an efficient way of visualising handwritten script signs based on their similarity. The image below to the left shows how the number one from the MNIST dataset has been clustered and visualised as small islands. The image to the right shows a heat map of each island and one can clearly see … Continue reading Creating an Atlas over Handwritten Script Signs
Making Large Collections of Handwritten Material Easily Accessible and Searchable
Libraries and cultural organisations have a rich amount of digitised historical handwritten material. A vast majority of this material has not yet been transcribed. The task of making these historical collections available for public access is challenging, especially in performing a simple text search across the collection. Machine learning based methods for handwritten text … Continue reading Making Large Collections of Handwritten Material Easily Accessible and Searchable
In Search of the Scribe Letter Spotting as a Tool for Identifying Scribes in Large Handwritten Text Corpora
In this article, letter spotting was used on a large set of handwritten documents in order to identify those that contain similar script. The main scribe of the Codex Holmiensis D 3 manuscript has yet not been identified in other documents. In the letter spotting process, a set of ‘g’:s, ‘h’:s and ‘k’:s have been selected … Continue reading In Search of the Scribe Letter Spotting as a Tool for Identifying Scribes in Large Handwritten Text Corpora
OptimalRANSAC
Due to its random nature, standard RANSAC is not always able to find the optimal set (all inliers) even for moderately contaminated sets and it is known to perform badly when the number of inliers is less than 50%. But this algorithm is capable of finding the optimal set (hence it is "almost" deterministic) even … Continue reading OptimalRANSAC
Clustering as an Alternative to RANSAC
These two papers propose clustering as a deterministic alternative to RANSAC. The corresponding pairs of points can be rewritten as points in 2D space. The inliers will be found close to each other while the outliers will be found further away from the cluster, depending on how well the points correspond. The first paper describes … Continue reading Clustering as an Alternative to RANSAC
Anders Hast 