Forum for Science, Industry and Business

Sponsored by:     3M 
Search our Site:


UC San Diego researchers give computers 'common sense'

Using a little-known Google Labs widget, computer scientists from UC San Diego and UCLA have brought common sense to an automated image labeling system. This common sense is the ability to use context to help identify objects in photographs.

For example, if a conventional automated object identifier has labeled a person, a tennis racket, a tennis court and a lemon in a photo, the new post-processing context check will re-label the lemon as a tennis ball.

“We think our paper is the first to bring external semantic context to the problem of object recognition,” said computer science professor Serge Belongie from UC San Diego.

The researchers show that the Google Labs tool called Google Sets can be used to provide external contextual information to automated object identifiers. The paper will be presented on Thursday 18 October 2007 at ICCV 2007 – the 11th IEEE International Conference on Computer Vision in Rio de Janeiro, Brazil.

Google Sets generates lists of related items or objects from just a few examples. If you type in John, Paul and George, it will return the words Ringo, Beatles and John Lennon. If you type “neon” and “argon” it will give you the rest of the noble gasses.

“In some ways, Google Sets is a proxy for common sense. In our paper, we showed that you can use this common sense to provide contextual information that improves the accuracy of automated image labeling systems,” said Belongie.

The image labeling system is a three step process. First, an automated system splits the image up into different regions through the process of image segmentation. In the photo above, image segmentation separates the person, the court, the racket and the yellow sphere.

Next, an automated system provides a ranked list of probable labels for each of these image regions.

Finally, the system adds a dose of context by processing all the different possible combinations of labels within the image and maximizing the contextual agreement among the labeled objects within each picture.

It is during this step that Google Sets can be used as a source of context that helps the system turn a lemon into a tennis ball. In this case, these “semantic context constraints” helped the system disambiguate between visually similar objects.

In another example, the researchers show that an object originally labeled as a cow is (correctly) re-labeled as a boat when the other objects in the image – sky, tree, building and water – are considered during the post-processing context step. In this case, the semantic context constraints helped to correct an entirely wrong image label. The context information came from the co-occurrence of object labels in the training sets rather than from Google Sets.

The computer scientists also highlight other advances they bring to automated object identification. First, instead of doing just one image segmentation, the researchers generated a collection of image segmentations and put together a shortlist of stable image segmentations. This increases the accuracy of the segmentation process and provides an implicit shape description for each of the image regions.

Second, the researchers ran their object categorization model on each of the segmentations, rather than on individual pixels. This dramatically reduced the computational demands on the object categorization model.

In the two sets of images that the researchers tested, the categorization results improved considerably with inclusion of context. For one image dataset, the average categorization accuracy increased more than 10 percent using the semantic context provided by Google Sets. In a second dataset, the average categorization accuracy improved by about 2 percent using the semantic context provided by Google Sets. The improvements were higher when the researchers gleaned context information from data on co-occurrence of object labels in the training data set for the object identifier.

Right now, the researchers are exploring ways to extend context beyond the presence of objects in the same image. For example, they want to make explicit use of absolute and relative geometric relationships between objects in an image – such as “above” or “inside” relationships. This would mean that if a person were sitting on top of an animal, the system would consider the animal to be more likely a horse than a dog.

Paper Information:

“Objects in Context,” by Andrew Rabinovich, Carolina Galleguillos, Eric Wiewiora and Serge Belongie from the Department of Computer Science and Engineering at the UCSD Jacobs School of Engineering. Andrea Vedaldi from the Department of Computer Science, UCLA.

Read that paper at:
The paper will be presented on Thursday 18 October 2007 at ICCV 2007 – the 11th IEEE International Conference on Computer Vision in Rio de Janeiro, Brazil.
National Science Foundation (NSF)
Afred P. Sloan Research Fellowship
Air Force Office of Scientific Research (AFOSR)
Office of Naval Research (ONR)
Author contact:
Serge Belongie
Media contact:
Daniel Kane

Daniel Kane | EurekAlert!
Further information:

More articles from Information Technology:

nachricht Next Generation Cryptography
20.03.2018 | Fraunhofer-Institut für Sichere Informationstechnologie SIT

nachricht TIB’s Visual Analytics Research Group to develop methods for person detection and visualisation
19.03.2018 | Technische Informationsbibliothek (TIB)

All articles from Information Technology >>>

The most recent press releases about innovation >>>

Die letzten 5 Focus-News des innovations-reports im Überblick:

Im Focus: Space observation with radar to secure Germany's space infrastructure

Satellites in near-Earth orbit are at risk due to the steady increase in space debris. But their mission in the areas of telecommunications, navigation or weather forecasts is essential for society. Fraunhofer FHR therefore develops radar-based systems which allow the detection, tracking and cataloging of even the smallest particles of debris. Satellite operators who have access to our data are in a better position to plan evasive maneuvers and prevent destructive collisions. From April, 25-29 2018, Fraunhofer FHR and its partners will exhibit the complementary radar systems TIRA and GESTRA as well as the latest radar techniques for space observation across three stands at the ILA Berlin.

The "traffic situation" in space is very tense: the Earth is currently being orbited not only by countless satellites but also by a large volume of space...

Im Focus: Researchers Discover New Anti-Cancer Protein

An international team of researchers has discovered a new anti-cancer protein. The protein, called LHPP, prevents the uncontrolled proliferation of cancer cells in the liver. The researchers led by Prof. Michael N. Hall from the Biozentrum, University of Basel, report in “Nature” that LHPP can also serve as a biomarker for the diagnosis and prognosis of liver cancer.

The incidence of liver cancer, also known as hepatocellular carcinoma, is steadily increasing. In the last twenty years, the number of cases has almost doubled...

Im Focus: Researchers at Fraunhofer monitor re-entry of Chinese space station Tiangong-1

In just a few weeks from now, the Chinese space station Tiangong-1 will re-enter the Earth's atmosphere where it will to a large extent burn up. It is possible that some debris will reach the Earth's surface. Tiangong-1 is orbiting the Earth uncontrolled at a speed of approx. 29,000 km/h.Currently the prognosis relating to the time of impact currently lies within a window of several days. The scientists at Fraunhofer FHR have already been monitoring Tiangong-1 for a number of weeks with their TIRA system, one of the most powerful space observation radars in the world, with a view to supporting the German Space Situational Awareness Center and the ESA with their re-entry forecasts.

Following the loss of radio contact with Tiangong-1 in 2016 and due to the low orbital height, it is now inevitable that the Chinese space station will...

Im Focus: Alliance „OLED Licht Forum“ – Key partner for OLED lighting solutions

Fraunhofer Institute for Organic Electronics, Electron Beam and Plasma Technology FEP, provider of research and development services for OLED lighting solutions, announces the founding of the “OLED Licht Forum” and presents latest OLED design and lighting solutions during light+building, from March 18th – 23rd, 2018 in Frankfurt a.M./Germany, at booth no. F91 in Hall 4.0.

They are united in their passion for OLED (organic light emitting diodes) lighting with all of its unique facets and application possibilities. Thus experts in...

Im Focus: Mars' oceans formed early, possibly aided by massive volcanic eruptions

Oceans formed before Tharsis and evolved together, shaping climate history of Mars

A new scenario seeking to explain how Mars' putative oceans came and went over the last 4 billion years implies that the oceans formed several hundred million...

All Focus news of the innovation-report >>>



Industry & Economy
Event News

New solar solutions for sustainable buildings and cities

23.03.2018 | Event News

Virtual reality conference comes to Reutlingen

19.03.2018 | Event News

Ultrafast Wireless and Chip Design at the DATE Conference in Dresden

16.03.2018 | Event News

Latest News

For graphite pellets, just add elbow grease

23.03.2018 | Materials Sciences

Unique communication strategy discovered in stem cell pathway controlling plant growth

23.03.2018 | Agricultural and Forestry Science

Sharpening the X-ray view of the nanocosm

23.03.2018 | Physics and Astronomy

Science & Research
Overview of more VideoLinks >>>