PRESS-NEWS.org - Press Release Distribution
PRESS RELEASES DISTRIBUTION

Carnegie Mellon researchers develop method to help computer vision systems decipher outdoor scenes

Carnegie Mellon researchers develop method to help computer vision systems decipher outdoor scenes
2010-09-09
(Press-News.org) PITTSBURGH—Computer vision systems can struggle to make sense of a single image, but a new method devised by computer scientists at Carnegie Mellon University enables computers to gain a deeper understanding of an image by reasoning about the physical constraints of the scene.

In much the same way that a child might use a set of toy building blocks to assemble something that looks like a building depicted on the cover of the toy set, the computer would analyze an outdoor scene by using virtual blocks to build a three-dimensional approximation of the image that makes sense based on volume and mass.

"When people look at a photo, they understand that the scene is geometrically constrained," said Abhinav Gupta, a post-doctoral fellow in CMU's Robotics Institute. "We know that buildings aren't infinitely thin, that most towers do not lean, and that heavy objects require support. It might not be possible to know the three-dimensional size and shape of all the objects in the photo, but we can narrow the possibilities. In the same way, if a computer can replicate an image, block by block, it can better understand the scene."

This novel approach to automated scene analysis could eventually be used to understand not only the objects in a scene, but the spaces in between them and what might lie behind areas obscured by objects in the foreground, said Alexei A. Efros, associate professor of robotics and computer science at CMU. That level of detail would be important, for instance, if a robot needed to plan a route where it might walk, he noted.

Gupta presented the research, which he conducted with Efros and Robotics Professor Martial Hebert, at the European Conference on Computer Vision, Sept. 5-11 in Crete, Greece.

Understanding outdoor scenes remains one of the great challenges of artificial intelligence. One approach has been to identify features of a scene, such as buildings, roads and cars, but this provides no understanding of the geometry of the scene, such as the location of walkable surfaces. Another approach, which Hebert and Efros pioneered with former student Derek Hoiem, now of the University of Illinois, Urbana-Champaign, has been to map the planar surfaces of an image to create a rough 3-D depiction of an image, similar to a pop-up book. But that approach can lead to depictions that are highly unlikely and sometimes physically impossible.

In the new method devised by Gupta, Efros and Hebert, the image is first broken into various segments corresponding to objects in the image. Once the ground and sky are identified, other segments are assigned potential geometric shapes. The shapes also are categorized as light or heavy, depending on appearance; a surface that appears to be a brick wall, for instance, would be classified as heavy.

The computer then attempts to reconstruct the image using the virtual blocks. If a heavy block appears unsupported, the computer must substitute an appropriately shaped block, or make assumptions that the original block was obscured in the original image.

Gupta said because this qualitative volumetric approach to scene understanding is so new, no established datasets or evaluation methodologies exist for it. He said in estimating the layout of surfaces, other than sky and ground, the method is better than 70 percent accurate, and its performance is almost as good when comparing its segmentation to ground truth. Overall, Gupta assesses the analysis as very good for 30 to 40 percent of the images and adequate for another 20 to 30 percent.



INFORMATION:

This work was supported by the National Science Foundation and a Guggenheim Fellowship to Efros. The Robotics Institute is part of Carnegie Mellon's School of Computer Science. Follow the school on Twitter @SCSatCMU.

About Carnegie Mellon University: Carnegie Mellon (www.cmu.edu) is a private, internationally ranked research university with programs in areas ranging from science, technology and business, to public policy, the humanities and the arts. More than 11,000 students in the university's seven schools and colleges benefit from a small student-to-faculty ratio and an education characterized by its focus on creating and implementing solutions for real problems, interdisciplinary collaboration and innovation. A global university, Carnegie Mellon's main campus in the United States is in Pittsburgh, Pa. It has campuses in California's Silicon Valley and Qatar, and programs in Asia, Australia, Europe and Mexico. The university is in the midst of a $1 billion fundraising campaign, titled "Inspire Innovation: The Campaign for Carnegie Mellon University," which aims to build its endowment, support faculty, students and innovative research, and enhance the physical campus with equipment and facility improvements.


[Attachments] See images for this press release:
Carnegie Mellon researchers develop method to help computer vision systems decipher outdoor scenes

ELSE PRESS RELEASES FROM THIS DATE:

High stress hormone levels linked to increased cardiovascular mortality

2010-09-09
Chevy Chase, MD—High levels of the stress hormone cortisol strongly predict cardiovascular death among both persons with and without pre-existing cardiovascular disease according to a new study accepted for publication in The Endocrine Society's Journal of Clinical Endocrinology & Metabolism (JCEM). In stressful situations, the body responds by producing the hormone cortisol. The effects of cortisol are intended to help the body recover from stress and regain a status of homeostasis, however chronically elevated cortisol levels have been associated with cardiovascular ...

Appetite hormones may predict weight regain after dieting

2010-09-09
Chevy Chase, MD—Many people have experienced the frustration that comes with regaining weight that was lost from dieting. According to a new study accepted for publication in The Endocrine Society's Journal of Clinical Endocrinology & Metabolism (JCEM), the levels of appetite hormones in the body prior to dieting may serve as a predictor of weight regain after dieting. "Treating obesity with drugs or dietary programs can be very effective in the short-term, but the long-term success of maintaining the weight lost is usually poor," said Ana Crujeiras, PhD, of Compejo Hospitalario ...

Drug-resistant malaria suggests a health policy change for pregnant women and infants

2010-09-09
September 9, 2010 – Malaria remains a serious global health problem, killing more than one million people per year. Treatment of the mosquito-borne illness relies on antibiotics, and the emergence of drug-resistant malaria is of growing concern. In a report published online today in Genome Research (www.genome.org), scientists analyzed the genomic features of a Peruvian parasite population, identifying the genetic basis for resistance to a common antibiotic and gaining new insights that could improve the efficacy of diagnosis and treatment strategies. The World Health ...

Researchers design more accurate method of determining premature infants' risk of illness

Researchers design more accurate method of determining premature infants risk of illness
2010-09-09
STANFORD, Calif. - Stanford University researchers have developed a revolutionary, non-invasive way of quickly predicting the future health of premature infants, an innovation that could better target specialized medical intervention and reduce health-care costs. "What the PhysiScore does is open a new frontier," said Anna Penn, MD, PhD, an assistant professor of pediatrics at the School of Medicine and a neonatologist at Lucile Packard Children's Hospital. "The national push toward electronic medical records helped us create a tool to detect patterns not readily ...

Researchers identify genes tied to deadliest ovarian cancers

2010-09-09
Scientists at the Johns Hopkins Kimmel Cancer Center have identified two genes whose mutations appear to be linked to ovarian clear cell carcinoma, one of the most aggressive forms of ovarian cancer. Clear cell carcinoma is generally resistant to standard therapy. In an article published online in the September 8 issue of Science Express, the researchers report that they found an average of 20 mutated genes per each ovarian clear cell cancer studied. Two of the genes were more commonly mutated among the samples: ARID1A, a gene whose product normally suppresses tumors; ...

Study identifies critical 'traffic engineer' of the nervous system

Study identifies critical traffic engineer of the nervous system
2010-09-09
Athens, GA—A new University of Georgia study published in the journal Nature has identified a critical enzyme that keeps traffic flowing in the right direction in the nervous system, and the finding could eventually lead to new treatments for conditions such as Alzheimer's and Parkinson's disease. "There was no medical or any other applied science drive for this project; it was purely curiosity about how transport inside cells works," said study co-author Jacek Gaertig, professor in the cellular biology department in the UGA Franklin College of Arts and Sciences. "But ...

Novel sensing mechanism discovered in dendritic cells to increase immune response to HIV

Novel sensing mechanism discovered in dendritic cells to increase immune response to HIV
2010-09-09
New York, NY (September 8, 2010) – Dendritic cells are the grand sentinels of the immune system, standing guard 24/7 to detect foreign invaders such as viruses and bacteria, and bring news of the invasion to other immune cells to marshal an attack. These sentinels, however, nearly always fail to respond adequately to HIV, the virus causing AIDS. Now a team of scientists at NYU Langone Medical Center has discovered a sensor in dendritic cells that recognizes HIV, spurring a more potent immune response by the sentinels to the virus. They report their findings in the September ...

House-sharing with microbes

2010-09-09
Household dust contains up to 1000 different species of microbes, with tens of millions of individual bacterial cells in each gram. And these are just the ones that can be grown in the lab! Dr Helena Rintala, speaking at the Society for General Microbiology's autumn meeting in Nottingham describes how we share our living and working spaces with millions of microbes, not all of whom are bad news. Microbes are a part of our normal environment and can be both beneficial and detrimental to our health. "Exposure to microbes in childhood can prevent the development of allergies. ...

Yeast holds clues to Parkinson's disease

2010-09-09
Yeast could be a powerful ally in the discovery of new therapeutic drugs to treat Parkinson's disease says a scientist presenting his work at the Society for General Microbiology's autumn meeting in Nottingham today. Dr Tiago Fleming Outeiro from the Instituto de Medicina Molecular in Lisbon, Portugal describes how his group is slowly uncovering the molecular basis of Parkinson's disease by studying the associated human protein in yeast cells. Parkinson's disease is a neurodegenerative disorder without any known cure that affects around 6 million people worldwide. The ...

Why chromosomes never tie their shoelaces

Why chromosomes never tie their shoelaces
2010-09-09
In the latest issue of the journal Nature, Miguel Godinho Ferreira, Principal Investigator at the Instituto Gulbenkian de Ciência (IGC) in Portugal, lead a team of researchers to shed light on a paradox that has puzzled biologists since the discovery of telomeres, the protective tips of chromosomes: while broken chromosome ends generated by DNA damage (such as radiation or cigarette smoke) are quickly joined together, telomeres are never tied to each other, thus allowing for the correct segregation of the genetic material into all cells in our body. Since telomeres erode ...

LAST 30 PRESS RELEASES:

Reality check: making indoor smartphone-based augmented reality work

Overthinking what you said? It’s your ‘lizard brain’ talking to newer, advanced parts of your brain

Black men — including transit workers — are targets for aggression on public transportation, study shows

Troubling spike in severe pregnancy-related complications for all ages in Illinois

Alcohol use identified by UTHealth Houston researchers as most common predictor of escalated cannabis vaping among youths in Texas

Need a landing pad for helicopter parenting? Frame tasks as learning

New MUSC Hollings Cancer Center research shows how Golgi stress affects T-cells' tumor-fighting ability

#16to365: New resources for year-round activism to end gender-based violence and strengthen bodily autonomy for all

Earliest fish-trapping facility in Central America discovered in Maya lowlands

São Paulo to host School on Disordered Systems

New insights into sleep uncover key mechanisms related to cognitive function

USC announces strategic collaboration with Autobahn Labs to accelerate drug discovery

Detroit health professionals urge the community to act and address the dangers of antimicrobial resistance

3D-printing advance mitigates three defects simultaneously for failure-free metal parts 

Ancient hot water on Mars points to habitable past: Curtin study

In Patagonia, more snow could protect glaciers from melt — but only if we curb greenhouse gas emissions soon

Simplicity is key to understanding and achieving goals

Caste differentiation in ants

Nutrition that aligns with guidelines during pregnancy may be associated with better infant growth outcomes, NIH study finds

New technology points to unexpected uses for snoRNA

Racial and ethnic variation in survival in early-onset colorectal cancer

Disparities by race and urbanicity in online health care facility reviews

Exploring factors affecting workers' acquisition of exercise habits using machine learning approaches

Nano-patterned copper oxide sensor for ultra-low hydrogen detection

Maintaining bridge safer; Digital sensing-based monitoring system

A novel approach for the composition design of high-entropy fluorite oxides with low thermal conductivity

A groundbreaking new approach to treating chronic abdominal pain

ECOG-ACRIN appoints seven researchers to scientific committee leadership positions

New model of neuronal circuit provides insight on eye movement

Cooking up a breakthrough: Penn engineers refine lipid nanoparticles for better mRNA therapies

[Press-News.org] Carnegie Mellon researchers develop method to help computer vision systems decipher outdoor scenes