Video eCommerce++: Towards Large Scale Online Video Advertising (Jun-2017)

Aim:

The main aim of our project is to exhibit appropriate product ads to particular users at proper timestamps of videos.

Proposed system:

The video advertisement only based on user preference. User preference find based on user behavior. When user will active his activities are observes, till user log out his account.User behavior analysis based on frequently search the product and how long user watches the product. User wants to buy a product, he spends more time to view the specification at the same he clicks the product more number of times to view the specification, price etc. These techniques used to find the user preference. After find a user preference, we embedded the user preference product’s advertisement in video in proper timestamp. Using surf Detector algorithm, Classify video frames based on products. Surf Detector is a algorithm, which extracts the some unique key points and descriptor from the product and frame and matches the key points to find the product in frame.