Research Article | Volume 2 Issue 2 - 2025
Compression and Decompression of Digital-Ink Handwriting Using Sparse Gaussian Process Regression and Dynamic Programming
Jinya Yano1 and Hiroyuki Fujioka2*
1Graduate School of Engineering, Fukuoka Institute of Technology, Japan
2Department of Information Management, Fukuoka Institute of Technology, Japan
*Corresponding Author: Hiroyuki Fujioka, Department of Information Management, Fukuoka Institute of Technology, 3-30-1 Wajiro-Higashi, Higashi-ku, Fukuoka 811-0295, Japan.
Abstract
This study addresses the challenges of compressing and decompressing digital-ink data captured from handwritten inputs on a device- such as pen-tablet, etc. We propose a novel method leveraging sparse Gaussian process regression (GPR) to compress digital-ink data into a compact kernel matrix using pseudo-inputs. To enhance compression accuracy and reconstruction fidelity, we integrate a dynamic programming (DP) strategy for optimal pseudo-input selection. Experimental results validate the proposed method by demonstrating its effectiveness and significant compression ratios.
Keywords: digital-ink; handwriting; compression; decompression; sparse Gaussian process regression; pseudo-inputs; optimal selection; dynamic programming
References
- G Kurtenbach. “Pen-based Computing, XRDS: Crossroads”. The ACM Magazine for Students-The Future of Interaction 16.4 (2010): 14-20.
- V Mazalov and S Watt. “Digital Ink Compression via Functional Approximation”. Proc. 12th Int. Conf. Frontiers in Handwriting Recognition, Kolkata, India (2010): 688-694.
- R Hu and S Watt. “Optimization of Point Selection on Digital Ink Curves”. Proc. 13th International Conference on Frontiers in Handwriting Recognition, Bari, Italy (2012): 525-530.
- Y Mieno, H Fujioka and H Kano. “Data Compression of Digital-Ink with Pen-Slips Using Multi-level L1 Smoothing Splines”. Proc. 2015 IEEE International Conference on Systems, Man, and Cybernetics, Hong Kong (2015): 1787-1792.
- R Tibshirani. “Regression Shrinkage and Selection via the Lasso”. J. of the Royal Statistical Society Series B 58.1 (1996): 267-288.
- SS Chen, DL Donoho and MA Saunders. “Atomic Decomposition by Basis Pursuit”. SIAM Review 43.1 (2001): 129-159.
- M Aharon, M Elad and A Bruckstein. “The K-SVD: An algorithm for designing of overcomplete dictionaries for sparse representations”. IEEE Trans. Image Process 54-11 (2006): 4311-4322.
- CM Bishop. “Pattern Recognition and Machine Learning”. Springer (2006).
- E. Snelson and Z. Ghahramani. “Sparse Gaussian Process using Pseudo-inputs”. Neural Information Processing Systems 13. MIT Press (2006).
- GH Golub and CF Van Loan. “Matrix Computations: Johns Hopkins Studies in Mathematical Sciences”. Johns Hopkins University Press. 3rd edition (1996).
- TA Davis. “Direct Methods for Sparse Linear Systems (Fundamentals of Algorithms, Series Number 2)”. Society for Industrial and Applied Mathematics (2006).
Citation
Hiroyuki Fujioka., et al. “Compression and Decompression of Digital-Ink Handwriting Using Sparse Gaussian Process Regression and Dynamic Programming". Clareus Scientific Science and Engineering 2.2 (2025): 03-11.
Copyright
© 2025 Hiroyuki Fujioka., et al. Licensee Clareus Scientific Publications. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license.
