AI software can identify objects in photos and videos at near-human levels

A new AI software program developed by researchers at Google and Stanford University can recognise objects in photos and videos at near-human levels of understanding.

ai software program google stanford university object recognition technology images videos human level understanding

It was only recently that computer systems became smart enough to identify unknown objects in photographs. Even then, it has generally been limited to individual objects. Now, two separate teams of researchers at Google and Stanford University have created software able to describe entire scenes. This could lead to much better and more intelligent algorithms in the future.

Stanford's work, entitled "Deep Visual-Semantic Alignments for Generating Image Descriptions", explains how specific details found in photographs and videos can be translated into written text. Google's version of the technology, in a study titled "Show and Tell: A Neural Image Caption Generator", produced similar results.

While each team used a slightly different approach, they both combined deep convolutional neural networks with recurrent neural networks that excel at text analysis and natural language processing. The programs were able to "learn" from each new interaction, with algorithms enabling the system to improve its accuracy by scanning scene after scene, looking for patterns, and then using the accumulation of previously described scenes to extrapolate what is being depicted in the next unknown image.

"The system can analyse an unknown image and explain it in words and phrases that make sense," says Fei-Fei Li, a professor of computer science and director of the Stanford Artificial Intelligence Lab. "This is an important milestone. It's the first time we've had a computer vision system that could tell a basic story about an unknown image by identifying discrete objects and also putting them into some context."

These latest algorithms are being trained on a visual dictionary – the ImageNet project – with a database of more than 14 million objects. Each object is described by a mathematical term, or vector, that enables the machine to recognise the shape the next time it is encountered. Those mathematical definitions are linked to the words humans would use to describe the objects.

“I was amazed that even with the small amount of training data that we were able to do so well,” said Oriol Vinyals, a Google computer scientist who worked with members of the Google Brain project. “The field is just starting, and we will see a lot of increases.”

In the near term, computer vision systems that can discern the story in a picture will enable people to search photo or video archives and find highly specific images. Eventually, these advances will lead to robotic systems able to navigate unknown situations. Driverless cars would also be made safer. However, it also raises the prospect of even greater levels of government surveillance.

"A group of young people playing a game of Frisbee."

"A person riding a motorcycle on a dirt road."

"A pizza sitting on top of a pan on top of a stove."

Comments

What is your preferred smartphone screen size

Screen envy. We all have it. Or do we? See, that’s the thing, I’m just not so sure. One thing I am sure about is that our phone screens have gotten progressively bigger and bigger over the last few years. The funny thing is, we as a society have taken that ball and run with it, like we just don’t care. But I always thought we did care? Some time ago, when phones were in the sub-5” range, we used to have a term called “phablet” which described an almost absurdly big phone. The definition varied from person to person. I always considered a phone with a screen 5.5” or bigger to be a phablet. The Galaxy Note fell into that category. The Lumia 1520 certainly did. There are others as well. I was safely ensconced in the 4.5” area at the time with my Lumia 920. What did I know? Leon’s getting laaaaaaaarrrger But slowly, phones started topping the 5” mark. I’m not talking about phablets here, I’m talking about normal flagship phones – the Galaxy S4 (barely) and the HTC One (M7) are ...

Best Web Design Company in Pondicherry

#Technology has two faces. We all feel it, but sometimes can’t find words to describe it. #Ebooks are the best example to show the 0-1 nature of emotions the #technology evokes. #itwhere provide a #Best #solutions to #Growyourbusiness feel free to drop a #Mail info@itwheretech.co.in www.itwheretech.co.in

Facebook Now Supports PGP To Send You Encrypted Emails

You can now instruct Facebook to encrypt every email it sends to you so nobody — not even the NSA — is likely to be able to read your messages anytime soon. All you have to do is import your public PGP key into your Facebook settings and you’re good to go. The problem here, of course, is that most people have no idea how public/private key email encryption works and how to even get started with it. In the wake of Edward Snowden’s leaks, a number of organizations, including Google, promised to completely hide the complexities of end-to-end email encryption from regular users. Very few of these products have materialized so far, however — not for lack of trying, but because this is actually a very complex problem, both from a technical and user experience perspective. Facebook uses the well-established PGP scheme (the GNU Privacy Guard implementation of OpenPGP, to be precise) to encrypt messages and tools lik...

eGym raises $45M Series C for cloud-connected gym equipment and fitness software

eGym , the Munich-based startup that offers cloud-connected gym equipment and supporting cloud software and app for the fitness training floor, has closed $45 million in Series C funding. The round was led by new investor HPE Growth Capital, while existing investors, including Highland Europe, also participated. The problem that eGym is looking to solve is that, whilst gyms have moved from a bodybuilder market to a mass market in the last 20 years, the technology in gyms lags behind. That’s despite the fact that better use of technology can help to reduce customer churn, the biggest pain-point of both gym operator and gym users. Comprising of an app for both gym user and trainer, combined with the company’s connected strength machines, the eGym Cloud makes it possible for gym members to receive better fitness instruction and an evolving and personalised fitness plan based on data collected as they workout. And by providing a better workout feedback loop, gym goers can get an i...

Lumia sales are in the toilet, Surface revenue is down

Microsoft has just released its Q1 FY2016 (Q3 CY2015) earnings report, posting revenue of $20.4 billion, operating income of $5.8 billion, net income of $4.6 billion and earnings per share of $0.57 (all GAAP figures). The software giant's numbers beat analysts' expectations, which has added around 10 percent to its stock price in after-hours trading. Here are the highlights of the software giant's quarter. Microsoft has divided its earnings in three categories, namely Productivity and Business Processes, Intelligent Cloud, and More Personal Computing. The good news comes from Intelligent Cloud, where revenue is up by eight percent, while the bad news is in More Personal Computing, where the poor performance of Lumia and Surface devices lead to a 17 percent decrease in revenue, year-over-year. Revenue from devices decreased by $1.8 billion -- or 49 percent, year-over-year -- thanks to a decrease of $1.5 billion in Phones revenue and a further $236 million in Surface ...

Exciting Technology

Search This Blog