Monday, Jun 25 2018 | Updated at 03:39 PM EDT

Stay Connected With Us F T R

Apr 05, 2017 12:10 PM EDT

Researchers at the University of Washington were able to demonstrate how easily Google's new AI tool for video searching can be deceived. The program uses machine learning to automatically analyze and label video content.

Google has recently released the Cloud Intelligence API which is a tool that helps developers to build apps that can automatically recognize objects and search for content within videos. With this, users can now search each moment of every video file in their catalog.

The tool is able to quickly annotate videos stored in Google Cloud Storage and helps users identify key nouns entities in their videos and when they occur in the footage. It also is able to separate signal from noise by getting relevant information at the video, shot or per frame.

TechXplore reported that UW electrical engineers and security researchers were able to show how the API can be deceived by slightly manipulating the videos. One can simply add subtle modifications to the video by adding an image to it so that the system returns only the labels related to the inserted image.

Watch video

In the study, they inserted an image of a car once every 50 frames into a video about animals. The system returned results that suggested the video was about an Audi.

This was the same team who found that Google's machine learning-based platform, which has been designed to identify and weed out comments from Internet trolls, can be easily deceived by typos. Misspelling offensive words or adding unnecessary punctuation can lead to the system ignoring the offensive content.

Radha Poovendran, chair of the UW electrical engineering department and director of the Network Security Lab and senior author of the study, said that machine learning systems are generally designed to give the best performance in ideal situations. However, in the real world, these systems are susceptible to intelligent subversion or attacks. He added that systems should be robust and resilient to adversaries as AI products are slowly being used in everyday applications.

See Now: Facebook will use AI to detect users with suicidal thoughts and prevent suicide

Follows uw, university of washington, Google, AI, Video Searching Tool, Cloud Intelligence API, API, Developer
© 2017 University Herald, All rights reserved. Do not reproduce without permission.

Must Read

Controlling Robots With Brainwaves And Hand Gestures

Jun 23, 2018 AM EDTWhat if you can control robots with a simple flick of a finger? A team of experts is trying to make that a reality through this latest experiment.

Flavored Electronic Cigarettes Linked To Possible Cardiovascular Disease

Jun 16, 2018 AM EDTScientists from the Boston University School of Medicine looked into the effects of flavored e-cigarettes to the lining of blood vessels. Here are ...

LendingTree Study: Which Places Have the Most Student Debt?

May 31, 2018 AM EDTLendingTree outs its study that identifies places in the United States with the most student debt. Here's the complete list.

Best College Reviews Names 10 Best Master's in Biomedical Engineering Programs Online

May 31, 2018 AM EDTPlanning to pursue a master's degree in Biomedical Engineering? Here are some of the best online programs you need to consider.