Synthesizing and Imitating Handwriting Using Deep Recurrent Neural Networks and Mixture Density Networks

Kumar, K. Suresh; Kandala, Harish; Reddy, Narsi

doi:10.1109/icccnt.2018.8493843

Cited by 13 publications

(6 citation statements)

References 1 publication

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Rico et al in [40] selected most prominent features by text rendering and applied distortion and curvature to transform these features in machine generated text to generate handwritten text. Kumar et al [41] trained LSTM for writer independent classifier using bernoulli and gaussian distributions to determine strokes and pen lifts in the various handwriting style texts.…”

Section: B Data Generation Methodsmentioning

confidence: 99%

Offline Hand Written Urdu Word Spotting Using Random Data Generation

et al. 2020

View full text Add to dashboard Cite

Urdu word spotting is among the most challenging tasks in image processing and word spotting of hand written Urdu text is even more so. When it comes to handwritten Urdu documents, variation among the same words of various writers is significant. The orientation and style of the handwriting makes it really challenging for a word spotting system to correctly recognize the instances of the keyword. In this research, we tend to overcome this hurdle. We propose a system that takes a database of hand written Urdu text and generates random, yet, similar images to improve the classifier's ability to recognize variations caused by difference in handwriting. For image generation, we used geometric transformations and variants of Generative Adversarial Network (GAN). For the word spotting process, Histogram of Oriented Gradients (HOG) features are extracted from ligature images and then used to train a Long Short-Term Memory (LSTM) network for the classification task. This is the first study that focuses on improving word spotting by generating arbitrary samples using GANs and its variants. The system achieved a promising recognition rate of 98.96% due to the sample generation using Cycle-GANs.

show abstract

Section: B Data Generation Methodsmentioning

confidence: 99%

Offline Hand Written Urdu Word Spotting Using Random Data Generation

et al. 2020

View full text Add to dashboard Cite

show abstract

“…For further end [9], of associating and separating the past information is troublesome. These difficulties prompts managing interrelated of information [5].…”

Section: Literature Workmentioning

confidence: 99%

Data Entities & its Privacy with Big Data Techniques in e-health systems

Khan,

Reddy,

Kumar*

2019

IJEAT

View full text Add to dashboard Cite

Enormous information is rising innovation now in different territories, i.e. like online purchase, online medical services, tweet investigation, and saving money area. Presently today's insurance agencies are appearing towards investigation of their tremendous data samples comprises of patients & healing center's data. From those informational indexes they extricating some valuable data. For the most part they focus on progress and disappointment rate and input provided by the patients. Patients are going to provide the hospital center bills alongside release outline, medicinal reports to the insurance agency. In point of patient system assurance agency is choosing to endorse case & recommend for newly coming patients. In this paper, patient reports, indications, log records & criticisms are dissected utilizing enormous information innovations like infinispan and guide lessen ideas for information extraction and isolation in E-medical coverage. Uncovering of patient’s confidential data has been finished utilizing confidential information encrypting calculation.

show abstract

“…Generative Regularized: Regularization relies upon shirking of overtraining through intentionally controlling the classifier's versatile quality [5]. Furthermore, theory presentations for a regularize classifier are better as they are all the more amazing to exemptions.…”

Section: Classification Algorithms For Eeg Based Bcimentioning

confidence: 99%

“…AAC is worn as an enhancement to reinstate vocalizations and/or material actions that is not serviceable. Currently, obtainable AAC technologies employ facial lexis, gestures, and symbols to produce voice and/or written output [5]. Individuals with extreme discourse or dialect issues depend on AAC to supplement existing discourse or supplant discourse that isn't practical.…”

Section: Augmentative and Alternative Communication (Aac)mentioning

confidence: 99%

Brain Waves Computation using ML in Gaming Consoles

Reddy¹,

Jyothsna²,

Swathi³

et al. 2019

IJRTE

View full text Add to dashboard Cite

ENCEPHALON ELECTRONIC BRAIN COMPUTING(EEBC) is a solution for a newer approach of establishing communication with the system and invasively controls devices by the user with his/her capabilities. BCI is a framework that does not stress physical activity, In fact develops a regulation by corroborating contact with human brain waves and takes these waves as commands to perform operations in system by using the method EEG (electroencephalogram) cap. Moreover, BCI terminology is suitable for substantiate involutions in hospital field, gaming field and fields capable in usage of brain waves. Physically disabled people with loss of functioning in muscular activity exhibit to use BCI technology to perform commands in gaming origins and evince in other basis of respective fields

show abstract

Synthesizing and Imitating Handwriting Using Deep Recurrent Neural Networks and Mixture Density Networks

Cited by 13 publications

References 1 publication

Offline Hand Written Urdu Word Spotting Using Random Data Generation

Offline Hand Written Urdu Word Spotting Using Random Data Generation

Data Entities & its Privacy with Big Data Techniques in e-health systems

Brain Waves Computation using ML in Gaming Consoles

Contact Info

Product

Resources

About