None of us can ever imagine browsing over the internet without images however those who are visually impaired make use of computers without any images. Although users are able to access the content by making use of screen reading technology, which offers written descriptions in order for blind people to interpret the images if there is any, some website Developers include descriptions of images in the cold in order to improve their website search engine rankings. In order to find out if the description is accurate or informative, what technology has been developed by Darren Guinness who is a PhD student in the Atlas interactive Robotics and Nova Technologies lab and the Superhuman Computing lab.
Visually impaired users who wish caption crawler to replace poor quality all text can press a keyboard shortcut requesting a replacement. The screen reader automatically speaks the new caption. The screen reader speaks the caption which is the longest caption found for a particular photo. Users can also make use of a different shot got to access any additional found captions.
Caption crawler only works with images which are used in multiple websites but the approach is effective as many website administrators offer informative for the descriptions. Caption crawler is a low cost and low latency solution, offering human quality captioning without incurring any additional cost in human labelling time.
Google chrome browser extension is used by Caption crawler in combination with a node.js cloud server. The caption crawler finds a caption for an image and it is streamed back to the browser extension.