Video Search: Still a Challenge

April 6, 2018

As MIT Technology Review describes in its article, “The Next Big Step for AI? Understanding Video,” artificial intelligence still tends to have trouble correctly interpreting video. A recent slew of new jobs at YouTube (owned by Google) underscores this flaw—“YouTube is Hiring 10,000 People to Police Offensive Videos,” reports the New York Post. When it comes to objectionable content, algorithms just don’t get it. Yet. Meanwhile, the PR machine keeps running.

MIT Tech editor Will Knight discusses some promising solutions in the above article, beginning close to home with a collaboration between MIT and IBM. He writes:

“MIT and IBM this week released a vast data set of video clips painstakingly annotated with details of the action being carried out. The Moments in Time Dataset includes three-second snippets of everything from fishing to break-dancing. ‘A lot of things in the world change from one second to the next,’ says Aude Oliva, a principal research scientist at MIT and one of the people behind the project. ‘If you want to understand why something is happening, motion gives you lot of information that you cannot capture in a single frame.’” … “The MIT-IBM project is in fact just one of several video data sets designed to spur progress in training machines to understand actions in the physical world. Last year, for example, Google released a set of eight million tagged YouTube videos called YouTube-8M. Facebook is developing an annotated data set of video actions called the Scenes, Actions, and Objects set.”

Knight also mentions Twenty Billion Neurons, which, he notes:

“… Created a custom data set by paying crowdsourced workers to perform simple tasks. One of the company’s cofounders, Roland Memisevic, says it also uses a neural network designed specifically to process temporal vision information.”

So, we should not be surprised if, soon, AI can comprehend what it “sees.” Meanwhile, sites that host video content would do well to employ the judgment of humans.

Cynthia Murrell, April 6, 2018

Written by Stephen E. Arnold · Filed Under News, Search, Video

Comments

Comments are closed.

Search the site
Subscribe to Beyond Search
Feature archive
News archive

Stephen E. Arnold monitors search, content processing, text mining and related topics from his high-tech nerve center in rural Kentucky. He tries to winnow the goose feathers from the giblets. He works with colleagues worldwide to make this Web log useful to those who want to go "beyond search". Contact him at sa [at] arnoldit.com. His Web site with additional information about search is arnoldit.com.