Clothing Genre Recognition System Using Image Processing Techniques- A Survey

Nowadays, Clothing business is one of the most important components in thee-commerce industry. So, there is plenty of online clothing sites are available where people can search and retrieve the most clothing items for their user query image. Clothing genre recognition is a very active topic in computer vision and multimedia research. In the textile industry, image processing techniques provide sensitive attention in the field of the image-based clothing recognition system. The sequence of cloth images can be given as input to the recognition system. This clothing genre recognition system helps to detect the patterns and features of cloths which helps to classify them using effective feature extraction and classification algorithms. Feature extraction techniques can be used to obtain features from the cloths. Classification algorithms from soft computing help to automatically classify clothes genres depending on style elements and their salient visual features. Deep learning and Support Vector Machine (SVM) classifier achieved better performance in classifying both upper wear and lower wear genres. The main motivations of this paper focus on automatically classifying both upper wear and lower wear genre from a full-body input image. Evaluation metrics like precision, recall, F-score were used to measure the classification accuracy. This paper addresses on issues, challenges, applications, frameworks, tools, and techniques for recognition of clothing genres is carried out.