 |
Conference papers
Unsupervised Neural Segmentation and Clustering for Unit Discovery in Sequential Data
Abstract : We study the problem of unsupervised segmentation and clustering of handwritten lines with applications to character discovery. We propose a constrained variant of Vector Quantized Variational Autoencoder (VQ-VAE) which produces a discrete and piecewise-constant encoding of the data. We show that the constrained quantization task is dual to a Markovian dynamics prior placed on the latent codes. Such view facilitates a probabilistic interpretation of the constraints and allows efficient inference. We demonstrate the effectiveness of the proposed method in the context of unsupervised handwriting character discovery in 17th-century scanned manuscripts.
Cited literature [14 references]
https://hal.archives-ouvertes.fr/hal-02399138
Contributor : Ricard Marxer <>
Submitted on : Sunday, December 8, 2019 - 9:57:22 PM
Last modification on : Sunday, November 22, 2020 - 3:08:14 AM
Long-term archiving on: : Monday, March 9, 2020 - 1:39:48 PM
Contributor : Ricard Marxer <>
Submitted on : Sunday, December 8, 2019 - 9:57:22 PM
Last modification on : Sunday, November 22, 2020 - 3:08:14 AM
Long-term archiving on: : Monday, March 9, 2020 - 1:39:48 PM
File
PGR009.pdf
Files produced by the author(s)