{"id":8758,"date":"2017-04-06T09:00:52","date_gmt":"2017-04-06T08:00:52","guid":{"rendered":"http:\/\/www.devopsonline.co.uk\/?p=8758"},"modified":"2017-04-04T11:32:01","modified_gmt":"2017-04-04T10:32:01","slug":"uber-uses-selfies-for-driver-facial-recognition","status":"publish","type":"post","link":"https:\/\/devopsnews.online\/uber-uses-selfies-for-driver-facial-recognition\/","title":{"rendered":"Uber uses selfies for driver facial recognition"},"content":{"rendered":"
Staying safe whilst in an Uber is one of the most important aspects of being able to ride in the California designed transportation network. One\u00a0aspect that differentiates Uber from its competitors is that customers are told beforehand on the app the drivers’ name, registration number, mobile number, a photograph and a tracking device so the customer can see exactly where the driver is.<\/p>\n
Uber is constantly updating its software to make sure the driver and client are who they say they are. Through engineering, the network travel company has taken a proactive approach with its new security solution called\u00a0Real-Time ID Check, which ensures the right person is behind the wheel.<\/p>\n
This feature protects riders from unverified drivers, and also prevents fraud by ensuring drivers\u2019 accounts are not compromised.<\/p>\n
Face verification stood out on top compared to all other solutions Uber technologies Inc. evaluated. A typical face verification algorithm involves three main steps:<\/p>\n
“Once we decided to use face verification, we explored a few leading vendors and ran a comparison analysis. We plotted each vendor\u2019s results as a ratio of the true positive rate (TPR) against the false positive rate (FPR) over varying degrees of match confidence thresholds,” Uber software engineers explained in the company blog<\/a>. \u00a0“The resulting receiver operating characteristic (ROC) curve allowed us to determine the maximum TPR possible while minimizing FPR. Looking for the solution\u00a0with the highest TPR, we ultimately chose to use Microsoft\u2019s Face API from its Cognitive Services suite.”<\/p>\n One of the key objectives for Real-Time ID Check was to avoid unnecessary friction for driver-partners. Therefore, Uber focused on making the user experience as seamless as possible throughout every stage of the project, as well as adding movement detection to ensure the randomly selected drivers asked to verify their identity aren’t prompted to take selfies whilst driving, but at the beginning of journeys instead.<\/p>\n The development\u00a0team refined designs and tweaked the engineering based on feedback from users and prototype testing. This enabled us to create a simple, yet effective experience that drivers can complete in only a few seconds.”This enabled us to create a simple, yet effective experience that drivers can complete in only a few seconds.<\/p>\n “This enabled us to create a simple, yet effective experience that drivers can complete in only a few seconds,” the engineers said.<\/p>\n <\/p>\n Edited from source by Ella Donaldson<\/p>\nTesting the user experience of the real-time ID check<\/h2>\n