Volume 18, No. 6, 2021
A Ml Approach Of Capturing, Augmenting And Utilizing Online Trends To Improve Textile Designs Patterns
S.Senthilvel , Dr. A.Prema
Abstract
Designing competitive design patterns in textile industry to improve sales is always a difficult creative task, where the designer knowledge need to be up-to-date with the latest trends and peoples preferences. This paper attempts a novel method of generating trendy design patterns with the help of supervised machine learning algorithms. A live dataset of ongoing trends will be collected from twitter hashtags through a customized tweepy API. The popularity of the images shared along with the hashtags are categorized through Content Based Image Retrieval (CBIR) technique. Various image augmentation techniques are used to generate different combinations of designs from the categorized pool of the popular images. Finally a Deep Neural architecture is developed to predict the best design patterns through supervised training. After implementation, a comparative analysis of various optimizations in neural architectures for developing popular patterns will be studied.
Pages: 4567-4574
Keywords: Social medias are the reflecting mirror of peoples trends in which terabytes of information are generated, shared and consumed by millions of people in daily basis.