Abstract | ||
---|---|---|
This paper considers a problem of highly compressing the data called "Digital-Ink" which is a sequence of position data sampled from the traced curve at a sampling rate. We here suppose that a set of digital-ink is measured and stored as two-dimensional position data by electronic device (e.g. smart-phone and pen-tablet PC, etc.). Then, we develop a method of digital-ink compression using B-spline approach with sparse coding. Such a compression method consists of two steps: (i) approximating by B-splines and (ii) sparse coding. By the step (i), the digital-ink is compressed as a control point vector which is a sequence of B-spline's control points. Then, in the step (ii), such a control point vector is shrunk to a sparse vector. In particular, employing a method of dictionary learning called 'K-SVD algorithm', we create the best dictionary so that the control point vector can be represented sparsely. We demonstrate the performance by some experimental studies. |
Year | Venue | Field |
---|---|---|
2016 | 2016 IEEE INTERNATIONAL CONFERENCE ON SYSTEMS, MAN, AND CYBERNETICS (SMC) | B-spline,Control point,K-SVD,Computer science,Neural coding,Sparse approximation,Artificial intelligence,Data compression,Machine learning,Sparse matrix,Encoding (memory) |
DocType | ISSN | Citations |
Conference | 1062-922X | 0 |
PageRank | References | Authors |
0.34 | 0 | 2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Hiroyuki Fujioka | 1 | 37 | 13.37 |
Hiroyuki Kano | 2 | 75 | 19.05 |