An Efficient Real time Product Recommendation using Facial Sentiment Analysis

Image recognition based on deep learning has made successful attempts in solving Computer Vision problems. Deep learning algorithms have primarily contributed to face detection, facial expression recognition, age and gender determination tasks. The performances of the algorithms have been evaluated by the researchers and they are made available as cloud services. By accessing the cloud, application developers can incorporate the already tested algorithms in their work and pay for the utilized service. Amazon Rekognition is a web service which provides highly accurate facial analysis and more suitable for facial recognition. Amazon Rekognition includes a simple, easy-to-use API that can quickly analyze any image file that's stored in Amazon S3. This paper explains the necessary steps to utilize the service to build an application for a retail store. The application captures the customer faces, performs facial analysis and recommends appropriate products by displaying the targeted advertisements. With the facial analysis responses of recognition algorithm, a decision block is devised to select the appropriate content to be displayed. The application is tested with anonymous images and performance is appreciable.