NEIL program labels images, learns associations with minimal help from people
A computer program called the Never Ending Image Learner (NEIL) is running 24 hours a day at Carnegie Mellon University, searching the Web for images, doing its best to understand them on its own and, as it builds a growing visual database, gathering common sense on a massive scale.
NEIL leverages recent advances in computer vision that enable computer programs to identify and label objects in images, to characterize scenes and to recognize attributes, such as colors, lighting and materials, all with a minimum of human supervision. In turn, the data it generates will further enhance the ability of computers to understand the visual world.
But NEIL also makes associations between these things to obtain common sense information that people just seem to know without ever saying — that cars often are found on roads, that buildings tend to be vertical and that ducks look sort of like geese. Based on text references, it might seem that the color associated with sheep is black, but people — and NEIL — nevertheless know that sheep typically are white.
“Images are the best way to learn visual properties,” said Abhinav Gupta, assistant research professor in Carnegie Mellon’s Robotics Institute. “Images also include a lot of common sense information about the world. People learn this by themselves and, with NEIL, we hope that computers will do so as well.”
A computer cluster has been running the NEIL program since late July and already has analyzed three million images, identifying 1,500 types of objects in half a million images and 1,200 types of scenes in hundreds of thousands of images. It has connected the dots to learn 2,500 associations from thousands of instances.
The public can now view NEIL’s findings at the project website, http://www.neil-kb.com.
The research team, including Xinlei Chen, a Ph.D. student in CMU’s Language Technologies Institute, and Abhinav Shrivastava, a Ph.D. student in robotics, will present its findings on Dec. 4 at the IEEE International Conference on Computer Vision in Sydney, Australia.
One motivation for the NEIL project is to create the world’s largest visual structured knowledge base, where objects, scenes, actions, attributes and contextual relationships are labeled and catalogued.
“What we have learned in the last 5-10 years of computer vision research is that the more data you have, the better computer vision becomes,” Gupta said.
Some projects, such as ImageNet and Visipedia, have tried to compile this structured data with human assistance. But the scale of the Internet is so vast — Facebook alone holds more than 200 billion images — that the only hope to analyze it all is to teach computers to do it largely by themselves.
Shrivastava said NEIL can sometimes make erroneous assumptions that compound mistakes, so people need to be part of the process. A Google Image search, for instance, might convince NEIL that “pink” is just the name of a singer, rather than a color.
“People don’t always know how or what to teach computers,” he observed. “But humans are good at telling computers when they are wrong.”
People also tell NEIL what categories of objects, scenes, etc., to search and analyze. But sometimes, what NEIL finds can surprise even the researchers. It can be anticipated, for instance, that a search for “apple” might return images of fruit as well as laptop computers. But Gupta and his landlubbing team had no idea that a search for F-18 would identify not only images of a fighter jet, but also of F18-class catamarans.
As its search proceeds, NEIL develops subcategories of objects – tricycles can be for kids, for adults and can be motorized, or cars come in a variety of brands and models. And it begins to notice associations – that zebras tend to be found in savannahs, for instance, and that stock trading floors are typically crowded.
The Latest Google Headlines on:
Machine Learning
[google_news title=”” keyword=”Machine Learning” num_posts=”10″ blurb_length=”0″ show_thumb=”left”]
The Latest Bing News on:
Machine Learning
- Why AI Won't Kill The Beauty Or Benefits Of Learning A Foreign Languageon April 27, 2024 at 9:09 am
Matsakis writes that the tool works so well she wonders if all her efforts at learning Mandarin were a waste.Automatic translation was not always so convincing. The early tools (Google Translate is ...
- Machine learning-powered robot streamlines genetic research processon April 26, 2024 at 9:26 pm
University of Minnesota Twin Cities researchers have constructed a robot that uses machine learning to fully automate a complicated microinjection process used in genetic research.
- Automated machine learning robot unlocks new potential for genetics researchon April 26, 2024 at 9:10 am
University of Minnesota Twin Cities researchers have constructed a robot that uses machine learning to fully automate a complicated microinjection process used in genetic research.
- 3 Machine Learning Stocks That Could Triple Your Money by 2030on April 26, 2024 at 3:58 am
InvestorPlace - Stock Market News, Stock Advice & Trading Tips Investing in machine learning (ML) stocks presents an enticing opportunity for ...
- The Top 3 Machine Learning Stocks to Buy in April 2024on April 26, 2024 at 3:15 am
InvestorPlace - Stock Market News, Stock Advice & Trading Tips The U.S. economy is poised for remarkable growth driven by advancements in ...
- Machine learning model predicts CIS to MS conversion risk: Studyon April 25, 2024 at 10:00 pm
A machine learning model can predict the risk of converting from clinically isolated syndrome (CIS) to multiple sclerosis (MS), per a study.
- Artificial intelligence, machine learning boost animal science researchon April 25, 2024 at 10:44 am
Science X is a network of high quality websites with most complete and comprehensive daily coverage of the full sweep of science, technology, and medicine news ...
- Machine learning and extended reality used to train welderson April 25, 2024 at 10:30 am
Ever since the ancient Egyptians hammered two pieces of gold together until they fused, the art of welding has continuously progressed.
- JFrog unveils MLflow integration to enhance machine learning model managementon April 25, 2024 at 6:15 am
Software supply chain company JFrog Ltd. today announced a new machine learning lifecycle integration between JFrog Artifactory and MLflow, an open-source software platform origin ...
- Machine learning and experimenton April 25, 2024 at 5:31 am
For more than 20 years in experimental particle physics and astrophysics, machine learning has been accelerating the pace of science, helping scientists tackle problems of greater and greater ...
The Latest Google Headlines on:
Never Ending Image Learner
[google_news title=”” keyword=”Never Ending Image Learner” num_posts=”10″ blurb_length=”0″ show_thumb=”left”]
The Latest Bing News on:
Never Ending Image Learnerb
- Overcoming Finnish language fears one step at a time — "If you don't speak, you won't learn"on April 26, 2024 at 6:17 am
Despite having a reputation for being very difficult, many immigrants to Finland find their own path to learning the Finnish language.
- Sophie Grégoire Trudeau is leaning into the unknown for her next chapter: 'I'm OK with the uncertainty'on April 23, 2024 at 4:00 am
No question was off limits in Yahoo Canada's candid and emotional conversation with the "Closer Together" author.
- Saudi Arabia to Reduce Cinema License Fees to Increase Economic Contributionon April 22, 2024 at 2:24 am
The Saudi government has presented a package of incentive programs to enhance the sustainability of the cinema sector, which includes reducing the financial fees for operating licenses for permanent ...
- Gulf Cinema Festival Concludes in Riyadhon April 20, 2024 at 10:27 am
The fourth edition of the Gulf Cinema Festival (GCF), organized by the Film Commission (FC), concluded in Riyadh under the patronage of Minister of Culture, Chairman of the Board of FC Prince Bader ...
- Charlie Munger's Investment Wisdom: 5 Timeless Lessons From Warren Buffett's Ex-Business Partner For Building Wealthon April 18, 2024 at 5:51 am
Former vice chairman of Berkshire Hathaway Charlie Munger was a long-time business partner of Warren Buffett. The two turned a small textile business into a staggering $860 billion empire with ...
- Machine pistol suspect now facing COVID-19 fraud charges for loanon April 18, 2024 at 2:30 am
WATERLOO — A Waterloo man accused of possessing a machine pistol is now facing fraud charges for COVID-19 relief money he received. Prosecutors with the U.S. District Court for Northern Iowa charged ...
- TDSB says 'New Deal' with province required to address budget challengeson April 17, 2024 at 2:14 pm
The Toronto District School Board says it requires a "New Deal" from the provincial government to address its funding challenges as it tries to balance its budget for the upcoming academic year.
- A Late Start Didn’t Slow Down Men’s Volleyball’s Jimmy Meinharton April 17, 2024 at 10:19 am
Going into his freshman year at Glenbard West High School, Meinhart’s passion for volleyball continued to grow, leading him to step away from baseball. His progression in high school allowed him to ...
- There’s Already a Solution to the STEM Crisis: It’s in High Schoolson April 17, 2024 at 12:15 am
Some solutions include creating a sense of joy and belonging in the classroom and making the high school schedule less rigid.
- Food Fight: How Corporations Ruined Food (2012) (trailer)on April 16, 2024 at 12:55 pm
When we walk into a supermarket, we assume that we have the widest possible choice of healthy foods. But in fact, over the course of the 20th century, our food system was co-opted by ...