INFORMATION:
How to spot deepfakes? Look at light reflection in the eyes
University at Buffalo deepfake spotting tool proves 94% effective with portrait-like photos, according to study
2021-03-12
(Press-News.org) BUFFALO, N.Y. - University at Buffalo computer scientists have developed a tool that automatically identifies deepfake photos by analyzing light reflections in the eyes.
The tool proved 94% effective in experiments described in a paper accepted at the IEEE International Conference on Acoustics, Speech and Signal Processing to be held in June in Toronto, Canada.
"The cornea is almost like a perfect semisphere and is very reflective," says the paper's lead author, Siwei Lyu, PhD, SUNY Empire Innovation Professor in the Department of Computer Science and Engineering. "So, anything that is coming to the eye with a light emitting from those sources will have an image on the cornea.
"The two eyes should have very similar reflective patterns because they're seeing the same thing. It's something that we typically don't typically notice when we look at a face," says Lyu, a multimedia and digital forensics expert who has testified before Congress.
The paper, "Exposing GAN-Generated Faces Using Inconsistent Corneal Specular Highlights," is available on the open access repository arXiv.
Co-authors are Shu Hu, a third-year computer science PhD student and research assistant in the Media Forensic Lab at UB, and Yuezun Li, PhD, a former senior research scientist at UB who is now a lecturer at the Ocean University of China's Center on Artificial Intelligence.
Tool maps face, examines tiny differences in eyes
When we look at something, the image of what we see is reflected in our eyes. In a real photo or video, the reflections on the eyes would generally appear to be the same shape and color.
However, most images generated by artificial intelligence - including generative adversary network (GAN) images - fail to accurately or consistently do this, possibly due to many photos combined to generate the fake image.
Lyu's tool exploits this shortcoming by spotting tiny deviations in reflected light in the eyes of deepfake images.
To conduct the experiments, the research team obtained real images from Flickr Faces-HQ, as well as fake images from http://www.thispersondoesnotexist.com, a repository of AI-generated faces that look lifelike but are indeed fake. All images were portrait-like (real people and fake people looking directly into the camera with good lighting) and 1,024 by 1,024 pixels.
The tool works by mapping out each face. It then examines the eyes, followed by the eyeballs and lastly the light reflected in each eyeball. It compares in incredible detail potential differences in shape, light intensity and other features of the reflected light.
'Deepfake-o-meter,' and commitment to fight deepfakes
While promising, Lyu's technique has limitations.
For one, you need a reflected source of light. Also, mismatched light reflections of the eyes can be fixed during editing of the image. Additionally, the technique looks only at the individual pixels reflected in the eyes - not the shape of the eye, the shapes within the eyes, or the nature of what's reflected in the eyes.
Finally, the technique compares the reflections within both eyes. If the subject is missing an eye, or the eye is not visible, the technique fails.
Lyu, who has researched machine learning and computer vision projects for over 20 years, previously proved that deepfake videos tend to have inconsistent or nonexistent blink rates for the video subjects.
In addition to testifying before Congress, he assisted Facebook in 2020 with its deepfake detection global challenge, and he helped create the "Deepfake-o-meter," an online resource to help the average person test to see if the video they've watched is, in fact, a deepfake.
He says identifying deepfakes is increasingly important, especially given the hyper-partisan world full of race-and gender-related tensions and the dangers of disinformation - particularly violence.
"Unfortunately, a big chunk of these kinds of fake videos were created for pornographic purposes, and that (caused) a lot of ... psychological damage to the victims," Lyu says. "There's also the potential political impact, the fake video showing politicians saying something or doing something that they're not supposed to do. That's bad."
ELSE PRESS RELEASES FROM THIS DATE:
Immuno-PET can give physicians early insight into tumor response to targeted therapy
2021-03-12
Reston, VA--Immuno-positron emission tomography (PET) imaging can provide early insight into a tumor's response to targeted therapy, allowing physicians to select the most effective treatment for patients who have cancer. The new research was published in the March issue of The Journal of Nuclear Medicine.
The research showed that immuno-PET successfully visualizes changes in different cancer receptors (receptor tyrosine kinases, or RTKs) within tumors during targeted therapies. This gives physicians a tool that can be used to evaluate the effectiveness of a treatment soon after its administration.
"When healthy cells turn into cancer cells, there is a disruption in the RTK signaling. This makes RTKs a valuable therapeutic and ...
CT colonography most effective noninvasive colorectal cancer screening test
2021-03-12
Leesburg, VA, March 12, 2021--According to an open-access article in ARRS' American Journal of Roentgenology (AJR), compared with multi-target stool-DNA (mt-sDNA) and fecal immunochemical test (FIT), CT colonography (CTC) with 10 mm threshold most effectively targets advanced neoplasia (AN)--preserving detection while decreasing unnecessary colonoscopies.
"CTC performed with a polyp size threshold for colonoscopy referral set at 10 mm represents the most effective and efficient non-invasive screening test for colorectal cancer (CRC) prevention and detection," clarified first author Perry J. Pickhardt from the department of radiology ...
New machine learning model could remove bias from social network connections
2021-03-12
UNIVERSITY PARK, Pa. -- Did you ever wonder how social networking applications like Facebook and LinkedIn make recommendations on the people you should friend or pages you should follow?
Behind the scenes are machine learning models that classify nodes based on the data they contain about users -- for example, their level of education, location or political affiliation. The models then use these classifications to recommend people and pages to each user. But there is significant bias in the recommendations made by these models -- known as graph neural networks (GNNs) ...
Study finds adolescents with autism may engage neural control systems differently
2021-03-12
A new study by UC Davis MIND Institute researchers suggests that executive control differences in autism spectrum disorder (ASD) may be the result of a unique approach, rather than an impairment.
Executive control difficulties are common in individuals with autism and are associated with challenges completing tasks and managing time. The study, published in Biological Psychiatry: Cognitive Neuroscience and Neuroimaging, sought to tease out whether these difficulties represent a disruption in proactive executive control (engaged and maintained before a ...
Use of perovskite will be a key feature of the next generation of electronic appliances
2021-03-12
Quantum dots are manmade nanoparticles of semiconducting material comprising only a few thousand atoms. Because of the small number of atoms, a quantum dot's properties lie between those of single atoms or molecules and bulk material with a huge number of atoms. By changing the nanoparticles' size and shape, it is possible to fine-tune their electronic and optical properties - how electrons bond and move through the material, and how light is absorbed and emitted by it.
Thanks to increasingly refined control of the nanoparticles' size and shape, the number ...
Artificial intelligence calculates suicide attempt risk
2021-03-12
A machine learning algorithm that predicts suicide attempt recently underwent a prospective trial at the institution where it was developed, Vanderbilt University Medical Center.
Over the 11 consecutive months concluding in April 2020, predictions ran silently in the background as adult patients were seen at VUMC. The algorithm, dubbed the Vanderbilt Suicide Attempt and Ideation Likelihood (VSAIL) model, uses routine information from electronic health records (EHRs) to calculate 30-day risk of return visits for suicide attempt, and, by extension, suicidal ideation.
Suicide has been on the rise in the U.S. for a generation ...
Association of acute symptoms of COVID-19, symptoms of depression in adults
2021-03-12
What The Study Did: Researchers investigated whether acute COVID-19 symptoms are associated with the probability of subsequent depressive symptoms.
Authors: Roy H. Perlis, M.D., M.Sc., of the Massachusetts General Hospital in Boston, is the corresponding author.
To access the embargoed study: Visit our For The Media website at this link https://media.jamanetwork.com/
(doi:10.1001/jamanetworkopen.2021.3223)
Editor's Note: Editor's Note: The article includes conflict of interest and funding/support disclosures. Please see the article for additional information, ...
Well-child visits with out-of-pocket costs before, after ACA
2021-03-12
What The Study Did: National claims data were used to look at changes in well-child care visits with out-of-pocket costs before and after passage of the Affordable Care Act.
Authors: Paul R. Shafer, Ph.D., of Boston University, is the corresponding author.
To access the embargoed study: Visit our For The Media website at this link https://media.jamanetwork.com/
(doi:10.1001/jamanetworkopen.2021.1248)
Editor's Note: Editor's Note: The article includes conflicts of interest disclosures. Please see the article for additional information, including other authors, author contributions and affiliations, conflict of interest and financial disclosures, and funding ...
Association between preterm birth, psychotropic drug use in adolescence, young adulthood
2021-03-12
What The Study Did: Researchers compared rates of psychotropic drug prescriptions during adolescence and young adulthood between individuals born preterm and at term.
Authors: Christine S. Bachmann, M.D., of the Norwegian University of Science and Technology in Trondheim, Norway, is the corresponding author.
To access the embargoed study: Visit our For The Media website at this link https://media.jamanetwork.com/
(doi:10.1001/jamanetworkopen.2021.1420)
Editor's Note: Editor's Note: The article includes funding/support disclosures. Please see the article for additional information, including other authors, ...
Poor survival after heart attack linked to excess levels of signaling protein in heart
2021-03-12
(Philadelphia, PA) - About 6.2 million Americans suffer from heart failure, an incurable disease with a staggering mortality rate - some 40 percent of patients die within five years of diagnosis. Heart failure is one form of heart disease, for which new therapies are desperately needed.
Now, in new work, scientists at the Lewis Katz School of Medicine (LKSOM) at Temple University identify a path to a promising novel therapeutic strategy, taking aim at a molecule in the heart known as G protein-coupled receptor kinase 5 (GRK5). In a study published online in the journal Cardiovascular Research, the scientists show in mice that reducing GRK5 levels can significantly improve survival ...
LAST 30 PRESS RELEASES:
Orphan nuclear receptors in metabolic dysfunction-associated steatotic liver disease development
A technological breakthrough for ultra-fast and greener AI
Pusan National University researchers identify key barriers hindering data-driven smart manufacturing adoption
Inking heterometallic nanosheets: A scalable breakthrough for coating, electronics, and electrocatalyst applications
Adults with autism show similar brain mapping of body parts as typically developing adults
Uncovering behavioral clues to childhood maltreatment
Premenstrual symptoms linked to increased risk of cardiovascular disease
Newly discovered remains of ancient river landscapes control ice flow in East Antarctica
Newly discovered interstellar object 'may be oldest comet ever seen'
Animal-inspired AI robot learns to navigate unfamiliar terrain
Underserved youth less likely to visit emergency department for concussion in Ontario, study finds
‘Molecular shield’ placed in the nose may soon treat common hay fever trigger
Beetles under climate stress lay larger male eggs: Wolbachia infection drives adaptive reproduction strategy in response to rising temperature and CO₂
Groundbreaking quantum study puts wave-particle duality to work
Weekly injection could be life changing for Parkinson’s patients
Toxic metals linked to impaired growth in infants in Guatemala
Being consistently physically active in adulthood linked to 30–40% lower risk of death
Nerve pain drug gabapentin linked to increased dementia, cognitive impairment risks
Children’s social care involvement common to nearly third of UK mums who died during perinatal period
‘Support, not judgement’: Study explores links between children’s social care involvement and maternal deaths
Ethnic minority and poorer children more likely to die in intensive care
Major progress in fertility preservation after treatment for cancer of the lymphatic system
Fewer complications after additional ultrasound in pregnant women who feel less fetal movement
Environmental impact of common pesticides seriously underestimated
The Milky Way could be teeming with more satellite galaxies than previously thought
New study reveals surprising reproductive secrets of a cricket-hunting parasitoid fly
Media Tip Sheet: Symposia at ESA2025
NSF CAREER Award will power UVA engineer’s research to improve drug purification
Tiny parasitoid flies show how early-life competition shapes adult success
New coating for glass promises energy-saving windows
[Press-News.org] How to spot deepfakes? Look at light reflection in the eyesUniversity at Buffalo deepfake spotting tool proves 94% effective with portrait-like photos, according to study