Date of Publication :22nd February 2018
Abstract: Augmented Reality is a new and emerging technology and many types of research and development are undergoing on it. Recent advancements in recognition of objects and characters present in an image or video have also come to an accuracy level which can be classified as reliable. In this paper, I propose an idea of implementation of different API’s to gather details about object present in the view. These objects may be a vehicle, buildings, book or a person. The whole process is broken down into 3 steps, pre-processing, object recognition and information gathering. Pre-processing is done to manipulate the image to help in object recognition such as changing the perspective or crop, object recognition is sending the image to API servers to recognise the object present in the view, and information gathering is to use the details about an object to find more details about it. For example, if there is a car in the view, pre-processing will modify the image, object recognition process will gather few details about it like make and model and then this information can be sent to another API to gather complete details about it
Reference :
-
- Carqueryapi.com (API documentation).
- Vehicles identification and premium options api reference, July 2015.
- Computer Vision, Microsoft Azure Documentation.
- BlipparAPI reference.