As Jacksonville City Council President Kevin Carrico got ready to nominate his boss for a spot on the JEA board, he sent the candidate a text message: “Guess it’s time they get a new board member to show them who’s boss.. … You ready to play the game?” The text was sent Jan. 20, three weeks before Paul Martinez, head of the Boys & Girls Club of Northeast Florida, was tapped for the utility’s board. The text was included in more that 1,000 documents Carrico delivered to the State Attorney’s Office in response to a subpoena on Feb. 24. Prosecutors sent the subpoena just days after a media report that Carrico had texted JEA board member Arthur Adams on Feb. 5 to say he would not be reappointed because the council president “owed a big favor to a friend and opted to put him of the JEA board.” That text also appeared in records delivered to the State Attorney’s Office. Jacksonville Today obtained Carrico’s response to prosecutors Tuesday, more than 100 days after requesting the documents from the city on Feb. 27. The disclosure followed dozens of emails and intervention from an attorney. Many of the 100 pages contain black-out information — or redactions — and a lack of context. But the material reveals new communications from Carrico to Martinez in the lead-up to the nomination and as media reports began to document it. Carrico officially nominated Martinez for the board seat Feb. 10, and the nomination was withdrawn at Martinez’s request on Feb. 24. Since late February, Carrico has accused JEA CEO Vickie Cavey of fostering a toxic workplace culture and made claims of racism at the utility’s corporate headquarters. He launched a City Council Special Investigatory Committee into JEA to investigate the largely anonymous claims. That led to
Jun 10, 2026 · via jaxtoday.org
A Fort Myers man is suing the Jacksonville Beach Police Department after he was wrongfully arrested almost two years ago due to a faulty facial recognition match, according to the American Civil Liberties Union of Florida. Robert Dillon, 52, sued the Jacksonville and Pinellas county sheriff’s offices and says police concealed evidence that shows he could not have committed the crime. Dillon was accused of trying to lure a child at a fast-food restaurant in Jacksonville Beach after the Jacksonville Sheriff’s Office ran grainy surveillance photos of the suspect through an AI-assisted facial recognition program, the suit says. The program, operated by the Pinellas Sheriff’s Office, identified Dillon as a possible match. A restaurant employee also picked Dillon’s photo out of a lineup, and Jacksonville Beach police arrested him in August 2024. “The night I spent in jail after they arrested me for a crime I did not commit still haunts me to this day,” Dillon said in a news release about the lawsuit. “I will never get over how terrified and worried I was, wondering if I’d ever go home to my wife and daughter again. Over a year later, I’m still picking up the pieces of my life, all because the police relied on this dangerous technology instead of doing their jobs and actually investigating.” The Jacksonville and Pinellas sheriff’s offices did not respond to Jacksonville Today‘s requests for comment Wednesday. Jacksonville Beach Police Department spokesperson Tonya Tator said the agency would not comment on the advice of its attorney. Private attorney Steve Silverberg, whose New York-based firm filed the lawsuit with the ACLU, said Dillon’s case illustrates what happens when police deploy AI-assisted identification tools without adequate safeguards. “Digital information can be a powerful tool for law enforcement, but its proliferation, supercharged by the AI boom, carries
Jun 10, 2026 · via jaxtoday.org
Florida Man Sues Police Over Wrongful Arrest Due to False Facial Recognition Match Robert Dillon, a long-time commercial crabber, was arrested for a crime he never committed in a city he’d never been to JACKSONVILLE, Fla. — Robert Dillon is suing the Jacksonville Beach Police Department, as well as the Jacksonville and Pinellas County Sheriff’s Offices, after he was wrongfully arrested due to a faulty facial recognition match. The police relied on an incorrect result from facial recognition technology to get an arrest warrant, while concealing evidence that showed he could not have committed the crime he was being accused of. He is one of 15 known people to have this happen to them in the United States. “The night I spent in jail after they arrested me for a crime I did not commit still haunts me to this day. I will never get over how terrified and worried I was, wondering if I’d ever go home to my wife and daughter again,” said Robert Dillon. “Over a year later, I’m still picking up the pieces of my life, all because the police relied on this dangerous technology instead of doing their jobs and actually investigating. Florida police must implement safeguards and ensure this never happens to anyone else, because until they do, nobody is safe.” Fifty-two-year-old Robert Dillon lives in Fort Myers, Florida with his wife and daughter. His life was irreparably changed in August 2024, when police arrested him for a crime he never committed. Mr. Dillon was accused of trying to lure a child at a fast-food restaurant in Jacksonville Beach, more than 300 miles away from his home, after a Jacksonville Sheriff’s Office employee ran grainy surveillance photos of the suspect through an AI-assisted facial recognition program, which identified Dillon as a possible match. Using
Jun 10, 2026 · via aclu.org
SUBSCRIBE TO OUR FREE NEWSLETTER Daily news & progressive opinion—funded by the people, not the corporations—delivered straight to your inbox. 5 #000000 #FFFFFF To donate by check, phone, or other method, see our More Ways to Give page. Daily news & progressive opinion—funded by the people, not the corporations—delivered straight to your inbox. "Over a year later, I'm still picking up the pieces of my life, all because the police relied on this dangerous technology instead of doing their jobs and actually investigating," said Robert Dillon. A federal lawsuit filed Wednesday in the Middle District of Florida by a Fort Myers resident wrongfully arrested nearly two years ago highlights the risks of police agencies relying on facial recognition tools. "This case is about what happens when police let an error-prone artificial intelligence (AI) system stand in for an investigation," explains the complaint, filed by attorneys with the state and national ACLU as well as the firm Hoguet Newman Regal & Kenney. "A facial recognition algorithm flagged Robert Dillon as the man who tried to lure or entice a child under 12 years old at a Jacksonville Beach McDonald's. It was wrong." The 52-year-old "lives more than 300 miles from" and "had never set foot in Jacksonville Beach," the complaint continues. "But rather than test the machine's answer against the evidence that would have cleared him, the officers built a case to confirm it. Mr. Dillon was arrested and prosecuted for one of the most stigmatizing crimes a person can face." Dillon—one of at least 15 people wrongfully arrested in the United States due to police reliance on incorrect facial recognition results—is suing the city of Jacksonville Beach as well as law enforcement officers from the Jacksonville Beach Police Department, Jacksonville Sheriff's Office (JSO), and Pinellas County Sheriff's Office. Reporting on
Jun 10, 2026 · via commondreams.org
This year, we're giving special recognition to some of our favorite products that truly set themselves apart. Wonderskin's Wonder Blading All-Day Lip Stain delivers stunning, daylong lip color from the ease of a doe-foot applicator, Essie's Ballet Slippers just might be the most timeless pale pink of all time, HigherDose's Infrared PEMF Pro Mat reduces inflammation and improves recovery, and Lumify's epic redness-reducing eye drops are a nonnegotiable in every editor's and makeup artist's getting-ready routine. Social Media Sensation Wonderskin Wonder Blading All-Day Lip Stain Wonderskin Wonder Blading All-Day Lip Stain You've likely seen Wonderskin's magical, shimmering blue lip stain all over the internet (it's also inspired many copycats), and we're here to say that this long-lasting lip stain is 100% worth the viral hype it's accrued over the last few years. Simply swipe this on dry lips as a liner or all over, wait, and wipe off. You'll be left with a gorgeous stain that lasts all day with zero upkeep. Why experts love it: "Aggressively worth the hype. I use it every day right after I brush my teeth. I do not leave the house without it on!" — Lila Childs, Makeup Artist "As we get older, we lose the color in our lips. I'd like to tell all the ladies before you get lip filler to try this lip stain! It makes your lips look plumper, and it doesn't dry them out. It brings back a youthful appearance that lasts all day." — Bridget Brager, Celebrity Hairstylist Is there a more iconic pale-pink nail polish? In our testers' opinions, no. We're excited to celebrate our favorite universally flattering ballet pink and the fact that it's remained a unanimous crowd-pleaser since its launch in 1982. You might be wondering if this nearly $1300 high-tech mat is worth the
Jun 10, 2026 · via whowhatwear.com
A Florida man is suing several law enforcement agencies for his arrest and prosecution for allegedly luring a child after he was wrongly identified using faulty AI facial recognition software. According to the Jacksonville Beach police department, an algorithm returned a 93% probability that Robert Dillon was the man caught on security cameras at a McDonald’s in the town attempting to persuade an unaccompanied girl, aged younger than 12, to leave with him. Dillon, however, lives in Fort Myers, more than 300 miles and a five-hour drive away, and told detectives he had never been to Jacksonville Beach in his life. The case was dismissed and charges dropped last year over the August 2024 incident. Now the 52-year-old has filed a lawsuit against the police department, the Jacksonville sheriff’s office, and Bob Gualtieri, the sheriff of Pinellas county, whose agency maintains and operates the Faces (Face Analysis Comparison and Examination) system and leases it to other law enforcement. “[The] investigation resulted in the wrongful arrest and prosecution of an innocent man,” the American Civil Liberties Union (ACLU) said in a lawsuit filed on Dillon’s behalf on Tuesday in district court in Fort Myers. “Mr Dillon was arrested at his home in front of his wife. He was accused of attempting to lure a child, a charge carrying devastating social stigma and permanent reputational destruction. He was subjected to months of criminal prosecution, and publicly branded with a mugshot that remains accessible online, long after the charges were dropped. “He no longer feels comfortable being friendly to children. No law enforcement agency has ever apologized or acknowledged the error.” The lawsuit further alleges that Dillon’s case is at least the 15th nationally to have involved a person being charged or arrested after a false identification. A Guardian investigation last month found
Jun 10, 2026 · via theguardian.com
Automated facial recognition tech led to a Black father being wrongfully imprisoned for months. Action News Jax reports that Jalil Richardson of Charlotte, NC, and father of 10, spent nearly three months behind bars for a car theft investigation that occurred in a Florida Publix parking lot. The case opened April 2, 2025, after an individual told the Jacksonville Sheriff’s Office that they had purchased a stolen car from a man and claimed they did not know it had been stolen. Richardson was matched to the crime by investigators who used automated facial recognition tech. They compared surveillance footage from the case to a photo of Richardson. The victim in the case also said Richardson was the suspect when presented with a lineup of photos, per Action News Jax. Richardson’s wife, Jasmine Jackson, said officers alleged Richardson was a near match to the suspect. “He said, ‘Jalil came back as close as 85%, and that’s the reason why he charged Jalil with the crime,’” Jackson told the outlet. Richardson was charged with “stolen property, grand theft, unlawful possession of motor vehicle with vehicle identification number removed, possession of fraudulent title, and fraudulent use of fictitious personal identification information,” according to Action News Jax. Richardson commented: “When they arrived, they informed me they had a warrant for me out in Jacksonville, and I was incarcerated for 33 days in Mecklenburg County,” Richardson said. “They extradited me to Florida after 33 days in Mecklenburg County, and then I spent the rest of the month in Jacksonville, which was another 50 days,” he said. Investigators later confirmed he did not commit the crime. Timecards proved that Richardson was working in North Carolina at the time of the crime in Jacksonville, FL, about 400 miles away. The case was dropped on May 27. “It’s
Jun 10, 2026 · via afrotech.com
The Computer Vision Pin takes question sout of the screen and into the physical world, turning concepts from image recognition technology into a collectible enamel accessory. The pin draws inspiration from artist Damjanski’s 2024 Berlin exhibition, which explored objects identified through computer vision systems. By shrinking those ideas into a wearable format, the design translates a technical process into a tangible artifact. Rather than functioning as a technology product, the pin acts as a conversation piece rooted in the intersection of art, artificial intelligence, and digital culture. Its aesthetic references the ways machines interpret and classify the visual world around them. The result is a niche collectible that appeals to developers, designers, artists, and technology enthusiasts who enjoy artifacts connected to computing culture. It offers a playful way to carry a concept from machine perception into everyday life. Tech-Inspired Art Merch Computer Vision Pin Transforms Machine Recognition Into Wearable Art Trend Themes - Algorithmic Aesthetics — Machine perception is becoming a source of visual language for collectible products that translate technical systems into culturally resonant design objects. - Wearable Tech-symbolism — Everyday accessories are evolving into identity markers for digital communities, blending subtle references to artificial intelligence, coding, and computational culture. - Exhibition-to-merch Collectibles — Gallery concepts are finding new commercial life through limited-edition merchandise that makes conceptual art more portable, affordable, and socially shareable. Industry Implications - Art Merchandise — Niche collectibles tied to digital culture are expanding the role of museum and artist merchandise beyond souvenirs into intellectual lifestyle products. - Fashion Accessories — Pins, badges, and small wearables are gaining relevance as compact canvases for technology-inspired self-expression among creative and technical audiences. - Artificial Intelligence — AI concepts are moving into consumer culture as symbolic artifacts, creating space for products that represent complex systems through approachable physical
Jun 10, 2026 · via trendhunter.com
eufy Security Launches Expanded FamiLock Smart Lock Lineup eufy FamiLock E40's AI-powered facial recognition technology delivers secure, hands-free home access BELLEVUE, WA, UNITED STATES, June 10, 2026 /EINPresswire.com/ -- eufy, a global leader in home security technology and smart appliances by Anker Innovations, today announced the launch of its extended FamiLock smart lock lineup, headlined by the new FamiLock E40, at select The Home Depot stores. Combining AI-powered facial recognition with privacy-first, on-device processing, the FamiLock lineup brings next-generation biometric home access technology to homeowners seeking secure, easy-to-install smart home upgrades.Built on eufyâs local-first data storage philosophy, the FamiLock lineup processes sensitive biometric data on-device, reinforcing user privacy with no monthly subscriptions attached*. Supporting the flagship FamiLock E40 are the FamiLock E35 and E32, extending the lineup with various configurations tailored to different household needs and price points. Combined together, the FamiLock lineup represents premium smart access solutions with advanced security and accessibility. eufy FamiLock E40 - Hands-free three-in-one video smart lock, camera, doorbell with industry-leading facial recognition Leading the lineup is the FamiLock E40, a premium smart lock that combines AI-powered facial recognition to create a truly hands-free entry experience. Designed for busy families, homeowners and frequent visitors, the biometric system enables fast and secure identity-based access. The FamiLock E40 also integrates a 2K video doorbell into its three-in-one video smart lock, camera, doorbell design, combining access control and monitoring in a single device. ⢠MSRP: $299.99 ⢠ANSI/BHMA certified for proven security and long-term durability, meeting high safety and performance requirements ⢠Multiple unlocking methods, including facial recognition, fingerprint recognition, app control, passcodes and physical keys ⢠Flexible family access for multi-user households, supporting up to 50 recognized faces and 50 fingerprints ⢠Provides real-time activity notifications and remote access management via the eufy App ⢠Local-first security
Jun 10, 2026 · via tech.einnews.com
Abstract Applications such as medical imaging, augmented and virtual reality, and embodied artificial intelligence (AI) depend on the ability to reconstruct complex signals from sparse observations. These applications are characterized by incomplete measurements and limited computational resources. Traditional approaches to digital hardware face the following challenges: explicit signal representations require heavy sampling and storage, data movement across the von Neumann bottleneck dominates energy and latency, and CMOS (complementary metal–oxide–semiconductor)-based circuits offer limited parallel efficiency. Here we present a software–hardware co-optimization framework for sparse-input signal reconstruction. At the software level, we use neural fields1 to implicitly represent signals using neural networks, which are further compressed by low-rank decomposition and structured pruning. At the hardware level, we design a resistive-memory-based computing-in-memory platform, featuring a Gaussian encoder and a multi-layer perceptron processing engine. The Gaussian encoder leverages the intrinsic stochasticity of resistive memory for efficient encoding, whereas the processing engine enables precise weight mapping through a hardware-aware quantization circuit. On a 40-nm 256 Kb resistive-memory macro, the system delivers 23.5×, 21.0× and 32.3× gains in projected energy efficiency, together with 10.8×, 38.8× and 6.2× gains in projected parallelism, for three-dimensional computed tomography sparse reconstruction, novel view synthesis and dynamic-scene novel view synthesis, without compromising on reconstruction quality. This work advances AI-driven signal reconstruction technology and paves the way for future efficient and robust medical AI and three-dimensional vision applications. This is a preview of subscription content, access via your institution Access options Access Nature and 54 other Nature Portfolio journals Get Nature+, our best-value online-access subscription $32.99 / 30 days cancel any time Subscribe to this journal Receive 52 print issues and online access $199.00 per year only $3.83 per issue Buy this article - Purchase on SpringerLink - Instant access to the full article PDF. USD 39.95 Prices may be subject to
Jun 10, 2026 · via nature.com
Department of Injustice Jalil Richardson of North Carolina is free after spending over 50 days in jail after being wrongfully arrested for a crime he did not commit. According to Action News Jax, Richardson was initially accused of stealing a vehicle in Jacksonville, Florida, after police fed surveillance video from a private business into their AI-integrated facial recognition system. The system then identified Richardson with what it said was an 85 percent facial recognition match, the Florida attorney’s office told Jax. Paired with two “eyewitness” accounts, it was enough to establish probable cause against Richardson, even though he’d been clocked into his job hundreds of miles away when the crime took place. After being arrested and made to spend nearly two months in custody, Richardson and his lawyers were finally able to establish his alibi in court, forcing prosecutors to drop the case — an infuriating miscarriage of justice, and quite possibly a sign of things to come as cops across the country embrace flawed facial recognition systems as a shortcut to investigating crimes. “There was no proper investigation done to even reach out to me or to see if I was even in Florida,” Richardson told Jax. “And I sat in there for over 50 days in the most worst jail ever.” Wrongful arrests based on AI facial recognition software are becoming something of a pattern with the Jacksonville Sheriff’s Office. Their first victim was Robert Dillon, a “93 percent match” who was wrongfully accused of attempting to lure and kidnap a 12-year-old child. Like Richardson, Dillon was a world away at the time — a five hour drive on the other side of the state. According to privacy litigation director for the Electronic Frontier Foundation Adam Schwartz, it’s the 14th known case of a wrongful arrest due to
Jun 10, 2026 · via futurism.com
AI software identifies robbery suspect ABILENE, Texas — A Taylor County man has been identified as the suspect in a series of armed robberies at convenience stores in Abilene earlier this month, according to a probable cause affidavit filed by law enforcement. Jacob Arriaga, 32, is accused of committing three robberies between June 2 and June 8 at convenience stores across the city while allegedly displaying or implying possession of a handgun and demanding cash from store clerks. RELATED| Automated license plate readers speed up Texas crime-fighting, police association says According to the affidavit, the first robbery occurred around 5:09 a.m. June 2 at the DK Convenience Store on South 14th Street. Investigators said a man entered the store under the pretense of needing change before brandishing what appeared to be a black 1911-style handgun and demanding money from the cash register. The suspect fled with cash and left in a newer blue four-door SUV. A second robbery was reported around 3:30 a.m. June 8 at another DK Convenience Store on East Highway 80. The affidavit states the suspect lifted his shirt to reveal what appeared to be a dark-colored handgun before demanding money. The clerk handed over approximately $125. RELATED| Five alleged gang members arrested after Abilene Police investigation About 10 minutes later, police responded to a robbery at the Allsups Convenience Store on North Judge Ely Boulevard. Investigators said the suspect waited outside until customers left before entering and asking for cigarettes. When the clerk turned to retrieve them, the suspect allegedly displayed a handgun and demanded cash. The clerk surrendered about $148. In each incident, witnesses described the suspect as a white male, about 5-foot-8, heavyset, with short brown hair and a full brown beard. Surveillance footage also showed the suspect wearing similar clothing during the robberies,
Jun 10, 2026 · via ktxs.com
MOSIP accredits Fime for biometric device testing French digital ID firm Fime has secured accreditation from MOSIP allowing it to conduct accredited biometric device testing for MOSIP-based digital identity programs. According to the company, the accreditation allows its laboratory to provide “independent, internationally recognized validation that biometric devices meet industry standards and MOSIP requirements.” National digital identity programs are sovereign infrastructure as vital as roads, hospitals, or schools, MOSIP’s Vice President for Partner Ecosystems, Sanjith Sundaram remarked after the accreditation. He added: “By accrediting Fime, MOSIP strengthens the ecosystem with independent testing capabilities that ensure biometric devices meet the highest standards of performance, security, and inclusivity.” Governments worldwide continue expanding national digital identity programs, with trust, inclusion and interoperability being among their priority considerations. Trust in digital ID lies largely on the quality of biometric data collected from subjects during enrollment, and this can only be guaranteed if devices deployed meet certain standards and specifications. Fime says the accreditation allows it to support governments in selecting biometric devices for MOSIP-based digital identity deployments during procurement. The accreditation also enables the company to test devices across population groups and deployment environments while supporting vendors seeking participation in the growing MOSIP ecosystem. Commenting on the accreditation, the SVP for Services at Fime, Noël Catherine, emphasized that biometric image quality is a key contributor to a trusted digital ID program. “With this accreditation, we support governments and partners with the validation needed to reduce risk, accelerate deployment, and deliver inclusive identity programs at scale,” she said. Fime’s laboratory is reputed for biometric testing across areas including matching performance, presentation attack detection, biometric data injection attack detection, and bias evaluation in biometric systems for populations of different age groups and characteristics. MOSIP’s growing role in national digital ID Biometric device testing for MOSIP falls
Jun 10, 2026 · via biometricupdate.com
About
Press
Copyright
Contact us
Creators
Advertise
Developers
Terms
Privacy
Policy & Safety
How YouTube works
Test new features
NFL Sunday Ticket
© 2026 Google LLC
Jun 9, 2026 · via youtube.com
Watch NASA's new experimental jet break the sound barrier for the first time NASA's X-59 jet has broken the sound barrier for the first time, a major milestone in the effort to build a quiet supersonic aircraft for civilians. The sleek, needle-nosed airplane exceeded the speed of sound for the first time on Friday, June 5. NASA test pilot Jim "Clue" Less took off in the plane at 11:08 a.m. PT and landed at Edwards Air Force Base in California about 81 minutes later. At the aircraft's top speed, it went Mach 1.1 — about 713 mph — at an altitude of 43,400 feet. The X-59 is experimental, part of the so-called QUESST mission to transform passenger air travel over land. Because existing supersonic aircraft produce loud sonic booms at high speed, the U.S. government bans routine supersonic flights over populated areas. But the X-59, designed by NASA and its contractor Lockheed Martin, is expected to tame the boom into a mere thump. You May Also Like NASA captured the achievement from the vantage point of a chase plane, which kept pace with the speedy experimental jet to monitor the test. You can watch the historic moment when it clocked supersonic speed in the video below. "X-59 goes through the number!" NASA administrator Jared Isaacman said on X. "We are rebuilding our X-plane portfolio and getting NASA back in the business of radical airframe and engine flight test!" An X‑plane is a U.S. aircraft designed to test new flight technologies and ideas. The goal of the X-59 is to provide regulators and the airline industry with the evidence needed to reconsider restrictions on supersonic aircraft. Most people think of NASA as the gateway to space, but the organization is first and foremost the nation's civil aeronautics agency (quite literally, the
Jun 9, 2026 · via mashable.com
By Raji Natarajan, Special to Rice News Picture a busy street during rush hour with vehicles zipping past and pedestrians rushing by. While humans can effortlessly perceive and interpret such dynamic scenes in real time, current 3D imaging technologies often struggle to create accurate representations of rapidly changing environments. A new study published in Nature Communications by researchers at Rice University and the University of Arizona introduces an innovative approach that could help machines see the world more clearly. Ashok Veeraraghavan, chair of the Department of Electrical and Computer Engineering in Rice’s George R. Brown School of Engineering and Computing, and Aniket Dashpute, a graduate student in his lab, in collaboration with associate professor Florian Willomitzer and his team at the University of Arizona’s Wyant College of Optical Sciences, developed a two-step computational imaging method that uses a laser and a high-speed camera to capture complex scenes in three dimensions with exceptional speed and accuracy. The key innovation lies in the system’s ability to transform ordinary matte surfaces into virtual screens. By using surrounding walls, furniture, clothing and other nonreflective surfaces as part of the imaging process, the technique enables accurate 3D reconstruction of scenes containing both matte and reflective objects, a long-standing challenge in computer vision. The advance could improve machine vision for applications ranging from industrial inspection and facial recognition to human sensing and autonomous vehicles. Most 3D imaging systems rely on structured light, which projects patterns onto a scene and measures how those patterns deform across object surfaces to create depth maps. While widely used, these systems can struggle with motion, challenging lighting conditions and scenes containing both matte and reflective materials. In mixed-reflectance environments, light bouncing between surfaces can distort measurements and reduce image quality. “To address these challenges, we leveraged a well-established technique in computer
Jun 9, 2026 · via news.rice.edu
Figures Abstract Accurate classification of sea turtle species is crucial for ecological monitoring and conservation, yet traditional visual classification methods remain limited by underwater imaging challenges such as occlusions, poor lighting, and background noise. To address these limitations, we propose an enhanced deep learning-based classification framework that integrates both color and structural features to improve the robustness of species recognition in complex marine environments. Building upon the ResNet-50 backbone, we introduce a four-channel input tensor comprising RGB data and Sobel-filtered edge maps, capturing both semantic and morphological information. Two novel fusion modules, LiteAFNet and AlphaBlendNet, are designed to integrate these features effectively. LiteAFNet leverages a lightweight attention mechanism to highlight discriminative regions, while AlphaBlendNet adaptively balances RGB and edge cues based on spatial context. Experimental results demonstrate significant improvements in classification performance across all evaluation metrics. Specifically, AlphaBlendNet achieves the highest precision (0.84), recall (0.88), F1-score (0.86), and mean average precision (mAP) of 87.2%, outperforming both the baseline fusion and LiteAFNet configurations. These results indicate that integrating color histograms with structural edge features enhances the model’s ability to distinguish between species with similar visual traits. This framework offers a scalable, accurate, and automated solution for underwater species classification and holds potential for broader application in marine biodiversity monitoring. Citation: Chaisiriprasert P, Deearom A (2026) A deep learning-based fusion framework for robust fine-grained classification of sea turtles in support of marine biodiversity. PLoS One 21(6): e0344942. https://doi.org/10.1371/journal.pone.0344942 Editor: Yaseen Ahmed Al-Mulla, Sultan Qaboos University, OMAN Received: August 13, 2025; Accepted: February 26, 2026; Published: June 9, 2026 Copyright: © 2026 Chaisiriprasert, Deearom. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Data Availability: DOIs 10.6084/m9.figshare.30452567 URLs
Jun 9, 2026 · via journals.plos.org
TL;DR Meta stripped NameTag facial recognition code from its AI app one day after WIRED exposed it on 50 million phones. Meta says no decision has been made. The NameTag system, which converted faces captured by Ray-Ban smart glasses into biometric signatures and stored unrecognised faces locally, had been embedded in the Meta AI app since January despite the company publicly saying it had made no final decision about facial recognition Meta stripped NameTag facial recognition code from its AI app one day after WIRED exposed it on 50 million phones. Meta says no decision has been made. Meta removed nearly all traces of an unreleased facial recognition system from its smart glasses companion app on Friday, one day after WIRED reported that the software had been quietly embedded in an app installed on more than 50 million phones. The feature, which Meta internally called NameTag, was designed to convert faces captured by the company’s Ray-Ban smart glasses into unique biometric signatures and compare them against a database stored on the user’s device. WIRED also found that faces the system failed to recognise were cropped, indexed, and stored locally for future processing. Andy Stone, Meta’s vice president of communications, told WIRED on Monday that the feature is “purely exploratory,” adding that no final decision has been made on what to do with it. That characterisation sits uneasily with the evidence WIRED documented. The version of Meta AI published the day of WIRED’s Thursday report contained several code libraries explicitly named for face recognition, a process for running the NameTag recognition pipeline, and a “Person recognised” alert the app would have shown if someone were identified. Friday’s release stripped all of it out, along with a folder where the app would have stored the cropped images and biometric signatures of unrecognised
Jun 9, 2026 · via thenextweb.com
Newswise — NEW YORK, 09 June 2026 – The IEEE Computer Society (CS) and the Computer Vision Foundation (CVF) announced the award-winning papers from the 2026 Conference on Computer Vision and Pattern Recognition (CVPR), recognizing outstanding achievements in computer vision. Best Paper Awards Following a rigorous review process that resulted in 4,089 accepted papers from 16,092 submissions, the CVPR 2026 Awards Selection Committee selected the following two papers for top honors at this year’s conference: CVPR 2026 Best Paper - Efficiently Reconstructing Dynamic Scenes One D4RT at a Time, Authors: Chuhan Zhang; Guillaume Le Moing; Skanda Koppula; Ignacio Rocco; Liliane Momeni; Junyu Xie; Shuyang Sun; Rahul Sukthankar; Joëlle K. Barral; Raia Hadsell; Zoubin Ghahramani; Andrew Zisserman; Junlin Zhang; Mehdi S. M. Sajjadi - A team from Google DeepMind, the University College London, and the University of Oxford developed D4RT, a network that can reconstruct the geometry and motion of dynamic 4D scenes from video. Using a unified transformer-based architecture, the model estimates depth, spatio-temporal correspondence, and full camera parameters, allowing for the independent and efficient probing of a 3D position of any point in space and time. By simplifying what has traditionally been a computationally intensive process, D4RT provides a lightweight and highly scalable method that enables remarkably efficient training and inference. CVPR 2026 Best Student Paper - Native and Compact Structured Latents for 3D Generation, Authors: Jianfeng Xiang; Xiaoxue Chen; Sicheng Xu; Ruicheng Wang; Zelong Lv; Yu Deng; Hongyuan Zhu; Yue Dong; Hao Zhao; Nicholas Jing Yuan; Jiaolong Yang - A team from Tsinghua University, Microsoft Research, the University of Science and Technology of China, and Microsoft AI developed a new approach to 3D generative modeling that significantly improves the quality and realism of AI-generated 3D assets. The research is centered around O-Voxel, a novel representation that can accurately
Jun 9, 2026 · via newswise.com
SPARC AI Demonstrates 43km Target Acquisition in GPS-Denied Test Event summary - SPARC AI completed a 43km target acquisition test over open water in Port Phillip Bay, Australia, at a drone height of 115m. - The test demonstrated capability comparable to the Strait of Hormuz, a critical maritime chokepoint. - SPARC AI integrated image recognition into its Overwatch drone controller application. - Overwatch now combines targets from multiple drones and manufacturers into a single operating map. - Next phase will introduce multi-drone deployment and swarm capabilities in GPS-denied environments. The big picture SPARC AI's successful long-range test and image recognition integration position Overwatch as a critical software layer for defense drones, particularly in GPS-denied environments. The ability to operate across multiple manufacturers expands its addressable market, challenging traditional proprietary drone platforms. The next phase of development, focusing on multi-drone coordination, could further solidify its competitive edge in autonomous defense systems. What we're watching - Market Expansion - Whether SPARC AI can sustain its position as a premium software layer across diverse drone manufacturers. - Technological Integration - The pace at which multi-drone deployment and swarm capabilities will be adopted by defense partners. - Geopolitical Demand - How the Strait of Hormuz comparison will influence adoption in contested maritime environments. Related topics
Jun 9, 2026 · via briefglance.com