Fashion Classification and Object Detection Using CNN

Shopkeepers once were the best choice for product details and were also guiding consumers to navigate the buying process. But technology has turned this scenario in reverse. With every minute of change in technology, information right from product specification to product analysis is all available online. With the help of practical policy of return on offer for E-commerce sites offering flexible return policies, it has become imperative for sellers to provide premium quality and customer service well. In this paper, we have demonstrated the effectiveness of computer vision for object detection methods. We have put forth the best method to get the dress type in the picture and give the exact name of that particular dress from the different journal papers. Here, we have used the VGG16 layer ConvNet architecture using multiple layers to get more effective results. In the proposed methodology, we came up with the few basic steps that are data preprocessing (Yamaguchi et al in 2012 IEEE conference on computer vision and pattern recognition, pp 3570–357, 2012 [1]), training (Lao and Jagadeesh in Convolutional neural networks for fashion classification and object detection, 2015 [2]), compilation, validation and model performance. The proposed method is experimentally tested, verified and found acceptable by achieving accuracy of 95%.