Contour Pattern Recognition with MNIST Dataset

Kaplun, Victoria; Shevlyakov, Artem N.

doi:10.1109/dynamics56256.2022.10014982

Cited by 3 publications

(2 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The two sets were collected from Census Bureau employees and high school students respectively [ 45 ]. This vast database of handwritten digits has been shown useful in pattern recognition and training various image processing systems for classification, with the aid of convolution neural network techniques [ 46 , 47 ]. Original images from MNIST were first being size-normalized, with the corresponding aspect ratio remaining unchanged, so that they could fit into a 20 × 20 pixel box; then, the center of mass of all pixels was computed, so that these processed MNIST images could be positioned at the centre of a “28 × 28 pixel grayscale image” [ 45 ].…”

Section: Flowchart and Data Sourcesmentioning

confidence: 99%

Application of Variational AutoEncoder (VAE) Model and Image Processing Approaches in Game Design

Mak

Han

Yin

2023

Sensors

View full text Add to dashboard Cite

In recent decades, the Variational AutoEncoder (VAE) model has shown good potential and capability in image generation and dimensionality reduction. The combination of VAE and various machine learning frameworks has also worked effectively in different daily life applications, however its possible use and effectiveness in modern game design has seldom been explored nor assessed. The use of its feature extractor for data clustering has also been minimally discussed in the literature neither. This study first attempts to explore different mathematical properties of the VAE model, in particular, the theoretical framework of the encoding and decoding processes, the possible achievable lower bound and loss functions of different applications; then applies the established VAE model to generate new game levels based on two well-known game settings; and to validate the effectiveness of its data clustering mechanism with the aid of the Modified National Institute of Standards and Technology (MNIST) database. Respective statistical metrics and assessments are also utilized to evaluate the performance of the proposed VAE model in aforementioned case studies. Based on the statistical and graphical results, several potential deficiencies, for example, difficulties in handling high-dimensional and vast datasets, as well as insufficient clarity of outputs are discussed; then measures of future enhancement, such as tokenization and the combination of VAE and GAN models, are also outlined. Hopefully, this can ultimately maximize the strengths and advantages of VAE for future game design tasks and relevant industrial missions.

show abstract

Section: Flowchart and Data Sourcesmentioning

confidence: 99%

Application of Variational AutoEncoder (VAE) Model and Image Processing Approaches in Game Design

Mak

Han

Yin

2023

Sensors

View full text Add to dashboard Cite

show abstract

“…The third type of dataset was extracted from the MNIST database, which was created in 1998. This vast database of handwritten digits has been shown useful in pattern recognition and training various image processing systems for classification, with the aid of convolution neural network techniques [43,44]. This database consists of 60,000 '28 × 28 grayscale images'-each of which consists of 10 digits (from 0 to 9, inclusive); along with a test set that consists of 10,000 images [45].…”

Section: Modified National Institute Of Standards and Technology (Mni...mentioning

confidence: 99%

Application of Variational AutoEncoder (VAE) Model and Image Processing Approaches in Game Design

Mak¹,

Han²,

Yin³

2023

Preprint

View full text Add to dashboard Cite

In recent decades, the Variational AutoEncoder (VAE) model has shown good potential and capabilities in image generation and dimensionality reduction. The combination of VAE and various machine learning frameworks has also worked effectively in different daily life applications, however its possibility and effectiveness in modern game design has seldom been explored nor assessed. The use of its feature extractor for data clustering was minimally discussed in literature neither. This paper first attempts to explore different mathematical properties of the VAE model, in particular, the theoretical framework of the encoding and decoding processes, the possible achievable lower bound and loss functions of different applications; then applies the established VAE model into generating new game levels within two well-known game settings; as well as validating the effectiveness of its data clustering mechanism with the aid of the Modified National Institute of Standards and Technology (MNIST) database. Respective statistical metrics and assessments were also utilized for evaluating the performance of the proposed VAE model in aforementioned case studies. Based on the statistical and spatial results, several potential drawbacks and future enhancement of the established model were outlined, with the aim of maximizing the strengths and advantages of VAE for future game design tasks and relevant industrial missions.

show abstract