Default Masking Scheme - Dates2026 - Ongoing - Author - Topics Landscape, Contemporary Issues, Nature & Environment, Fine Art - Location Shanghai, China This series begins with my own photographs of landscapes and examines how automated recognition transforms them into new images. By reworking the masks generated during the recognition process through layering, softening, and recomposition, I create lands I have always wanted to preserve the initial attraction of looking at an image. Before viewing turns into analysis, these works appear simply as seductive, ambiguous, and flattened landscapes. Yet when traced back to their making, that attraction emerges from a working process built upon automated image recognition. In this series, I explore how automated recognition systems interpret landscapes, and how that interpretation reshapes photographic images. Every work begins with my own photographs as source material, alongside masks generated through Camera Raw's automatic landscape recognition during the editing process. My process starts with these automatically generated masks. Rather than preserving the clear boundaries produced by the algorithm, I repeatedly overlay, soften, and reorganise them, transforming the traces of recognition into new image structures. Instead of presenting the software's workflow as an end in itself, I try to retain the images' visual appeal while searching for a balance between intuitive viewing and an awareness of their constructed nature. The red overlay is derived from the default colour used for masking, which itself originates from the Rubylith masking film historically used in printing and plate-making. Once a temporary and largely invisible stage within image production, this colour is reintroduced as part of the final image. It preserves traces of selection while making an otherwise hidden working process visible again, connecting contemporary digital recognition with earlier methods of image production. I hope these works are encountered first as images rather than understood immediately as
Aug 2, 2026 · via phmuseum.com
A wooden box bolted to a platform in a Costa Rican forest can spot a capuchin monkey’s face and choose a puzzle suited to that individual. Get the answer right, and a slice of banana pops out. Nobody has to stand there and run the test. Wild primates are usually studied one of two ways: in controlled laboratory settings or watched loose in the trees with almost no experimental control. Researchers at Emory University and the Georgia Institute of Technology built a new system to close that gap. Marcela Benítez, an assistant professor of anthropology at Emory, led the project, which the team calls CapuchinAI. She has spent years studying monkey behavior both in captivity and in the wild, and she wanted a way to bring lab-grade precision into the forest itself. “The primate brain didn’t evolve in a lab, it evolved in complex, competitive environments,” said Benítez. “Yet primate cognition is rarely studied in the wild because the experimental control needed to measure cognition is difficult in unpredictable environments.” The team first needed a computer that could tell one capuchin from another. They adapted an open-source facial-recognition program called YOLO. Then they trained it on thousands of close-up photos and videos of six adult male capuchins living at the Taboga Forest Reserve. Undergraduates at both universities spent hours drawing boxes around each monkey’s face in old field footage and tagging every image with the correct name. The team tested the finished model on pictures it had never seen, some blurred or lit differently than the training photos. It detected a capuchin’s face in the frame 98 percent of the time, and when it did, it identified the correct monkey 97 percent of the time. Federico Sánchez Vargas, an Emory PhD student in anthropology and the paper’s first author, took
Aug 2, 2026 · via earth.com
AI Rail Station Monitoring Could Move Beyond Passive CCTV 02.08.2026 INTERNATIONAL: AI rail station monitoring could shift security teams from continuous manual viewing towards connected camera, sensor and analytics networks that flag unattended objects and unusual crowd movement in real time. Rail stations combine heavy passenger flows with multiple entrances and open access. That design supports circulation, but it also makes perimeter control difficult and allows a person involved in an incident to blend into a crowd or leave quickly. The pressure is not limited to passenger theft. Security teams may also face antisocial behaviour, attacks on staff, equipment theft and threats to critical infrastructure. In Britain, reported attacks on station staff and police increased by 10% from 2019 despite an 11% reduction in passenger numbers, according to figures reported by the BBC. Connected monitoring replaces isolated camera feeds Conventional CCTV networks often operate as separate systems and depend on staff watching numerous feeds. Older cameras can also deliver images that are difficult to use when teams need to identify people, objects or the sequence of an incident. An IP-based architecture changes the operating model. Cameras can share data with sensors and access-control devices, while a video management system, or VMS, brings feeds and alerts into one interface. Instead of treating every frame equally, analytics can direct attention to events that match configured risk indicators. The applications described in a recent rail station monitoring analysis include unattended-object detection and assessment of crowd behaviour. Software can highlight loitering, an unexpected build-up of people or movement that differs from normal patterns. The same pattern-recognition approach can also support fault reporting and maintenance scheduling. This does not remove the operational role of security staff. It changes the first stage of monitoring: software identifies a possible event, while trained personnel assess the context and
Aug 2, 2026 · via railway.supply
As artificial intelligence becomes more capable, questions about whether machines can think, reason or even replace human intelligence have become increasingly common. Yet decades before AI entered everyday life, B.F. Skinner offered a strikingly different perspective. He wrote, "The real problem is not whether machines think but whether men do." Skinner did not doubt the capacities of the machine; he made us think about ourselves and our ways of thinking. Skinnerâs words suggest that the worst thing is not intelligent machinery but the danger that people will cease to question, analyse and think for themselves. B.F. Skinner (1904-1990) was an American psychologist, behaviourist and writer who made important contributions in the field of learning, behaviour and conditioning that have been influential in the areas of psychology, education and behavioural science. He was always interested in exploring the factors influencing human behaviour through experience and environment. Quote of the day by B.F. Skinner "The real problem is not whether machines think but whether men do." What does this quote mean? This quotation emphasises the fact that what really matters for the progress of society is not whether the machines get smarter, but whether humans keep on using their intelligence effectively. Technology is capable of processing information extremely fast, yet it is unable to substitute human intellect, curiosity, imagination, and moral values. Technology is only as wise as the people who use it Any significant technological revolution comes with tremendous opportunities and new obligations. The computer, Internet, and artificial intelligence are capable of resolving problems and being efficient, yet incapable of choosing between right and wrong or good for society. The decision rests on human evaluation. As Skinner points out, tools gain their meaning with proper judgment. Critical thinking matters more than ever As we live in a world where information is
Aug 2, 2026 · via timesofindia.indiatimes.com
For most of democratic history, the state could see a crowd but not remember it. A police officer watching a rally of 50,000 people could identify only a handful of faces and retain almost none of them by evening. This practical anonymity allowed citizens to participate in public life without fearing that their presence would become a permanent record. Artificial intelligence has begun to dismantle that arrangement. During the recent youth-led demonstrations organised by the Cockroach Janta Party (CJP) at Delhi's Jantar Mantar, the Delhi Police deployed live facial recognition systems mounted on mobile command vehicles. The deployment has since become the subject of a Public Interest Litigation before the Delhi High Court, raising a significant constitutional question: Can the freedoms of assembly and dissent survive when public presence itself becomes data? The controversy is not merely about cameras. Cameras have watched Indian crowds for decades. The fundamental transformation lies elsewhere—it is in the state's newfound ability to remember. A face captured today can potentially become an identity stored, matched and retrieved years later. How does facial recognition technology work? Facial recognition is often described as a smarter camera. That description is misleading. It is a biometric identification system that converts human features into a digital signature. When a camera captures a face, algorithms analyse facial landmarks—the distance between the eyes, the contour of the jaw, the depth of the nose bridge—and transform them into a numerical template known as a ‘faceprint’. The system does not merely record. It interrogates. It continuously asks whether a face corresponds to an identity already present in a database. Connected with police records and surveillance networks, facial recognition can scan thousands of individuals within minutes, without their knowledge or consent. Unlike fingerprints or iris scans, it requires neither cooperation nor physical contact. Presence in
Aug 1, 2026 · via newindianexpress.com
Microsoft's Windows installation files are getting noticeably larger, and the culprit isn't bloatware or legacy code - it's AI. The company's ISO downloads have been steadily creeping upward in size as on-device AI features require increasingly large language models and neural processing capabilities to be baked directly into the operating system. For users with smaller drives or limited bandwidth, this shift is creating real headaches. Microsoft users downloading fresh Windows installations are noticing something unusual - the ISO files keep getting bigger. What used to be a manageable download has ballooned into multi-gigabyte packages that strain both storage and bandwidth, particularly for those working with smaller drives or slower connections. The reason isn't the usual suspects of accumulated updates or legacy compatibility files. Instead, it's the march of AI directly into the operating system. As Microsoft pushes forward with AI-powered features across Windows, those capabilities need to live somewhere. Large language models, neural processing engines, and on-device AI assistants all require substantial storage footprints before they can run locally without constant cloud connectivity. This isn't just about Copilot or voice assistants. Modern Windows builds are increasingly shipping with AI models for everything from image recognition and real-time translation to predictive text and automated system optimization. Each of these features demands its own set of trained models, and those models take up space - sometimes hundreds of megabytes or more per feature. The shift represents a fundamental change in how operating systems are architected. Where previous Windows versions could offload intelligent features to cloud services, the industry is moving toward on-device AI that prioritizes privacy, speed, and offline functionality. But that convenience comes with a storage cost that gets passed directly to users during installation. For enterprise deployments, the implications are particularly significant. IT departments managing fleets of devices need to
Aug 1, 2026 · via techbuzz.ai
Getting your Trinity Audio player ready...
Highlights and lowlights from the week's news
Already have an account? Log In
This article is only available to subscribers
Offer valid for non-subscribers only
Aug 1, 2026 · via chicoer.com
The United States Mission in Nigeria has warned intending travellers against using artificial intelligence or digital editing tools on their passport photographs. The mission said such photos would not be accepted for visa applications. In a statement on its official X handle on Saturday, the mission said passport photos must be recent and must clearly resemble the applicant. It added that submitting an altered image would delay the application process. “Do NOT use AI or digital editing tools on your passport photo. We will not accept photos that are edited or filtered. “Your photo should be recent (taken within the past six months) and look like you (the TSA or CBP agent must be able to tell it’s you). Submitting a digitally enhanced photo will significantly delay your application,” the mission said. The warning was accompanied by a flier reiterating the policy in bold terms. “Do not use AI or digital editing tools on your passport photo. The Department of State will not accept photos that are edited or filtered. “Using a digitally enhanced photo will significantly delay your application,” the flier read. The mission directed applicants to consult the full photo requirements on the State Department’s website, travel.state.gov/photo. The advisory comes amid a broader push by the US authorities to curb the use of AI-manipulated images in official documentation, as facial recognition systems used by border agencies increasingly rely on unaltered, high-fidelity photographs to verify travellers’ identities.
Aug 1, 2026 · via punchng.com
Windows Hello vs. Enhanced Sign‑in Security: Which sign‑in method actually keeps your Windows 11 PC safer, and what's the difference? Windows 11's latest security update expands Enhanced Sign-in Security to external fingerprint readers. Here's what it actually changes. Microsoft is expanding one of the least understood security features in Windows 11. Beginning with the August 2026 update, Enhanced Sign-in Security (ESS) now supports compatible external fingerprint readers, extending the company's most secure Windows Hello experience to devices without built-in biometric hardware. The timing couldn't be better because Enhanced Sign-in Security has confused users ever since Microsoft introduced it. Some users assume it's simply a newer version of Windows Hello, while others think it's reserved for businesses or Copilot+ PCs. The reality is that Windows Hello and Windows Hello Enhanced Sign-in Security use the same sign-in experience, but they protect your biometric data in different ways. After spending time digging through Microsoft's support page, I think that's the distinction most explanations miss. Enhanced Sign-in Security isn't about making facial recognition more accurate or fingerprint sign-in faster. It's about making the entire authentication process harder to attack. Standard Windows Hello is already one of the best security features Before the Enhanced Sign-in Security feature was introduced, Windows Hello had already replaced passwords with a much stronger authentication model. Instead of storing passwords that can be stolen or reused, Windows Hello creates cryptographic credentials that are attached to the Trusted Platform Module (TPM) available on your computer. Facial recognition and fingerprints are used only to unlock those credentials, and your biometric templates remain on the device rather than being uploaded to Microsoft's servers. For the average home user, that already provides excellent protection against phishing, password reuse, and stolen credentials. The Enhanced Sign-in Security feature doesn't replace the architecture already in place. Instead,
Aug 1, 2026 · via windowscentral.com
Gay bars in San Franscisco are installing ID and facial scanners and requiring patrons to give it up if they want to get in, with Cydney Hayes in The San Francisco Gazetteer likening the gates to the TSA and Jacob Ogles in The Advocate pointing out the loss of anonymity and safety in LGBTQ+ spaces. Here's Hayes: Owen had just handed her ID to the bouncer when she was directed to turn and face a small camera before being allowed into the bar. "I was just kind of taken aback," Owen recalled. The camera was hooked up to a monitor and a forensic ID scanner. It looked like the sort of high-security kiosk you'd find while going through airport security. "Why is this at a gay bar, of all places?" At least three Castro bars (Mix, Badlands and Toad Hall) have deployed these PatronScan Guard+ kiosks (previously at Boing Boing), which scan IDs and photograph patrons at the door. The devices record names, birthdays, photos, genders and ZIP codes as well as when you came in. The data is retained for "up to five years" and shared across the network of venues using the system. Bouncers are not required to tell patrons they are being photographed. The Electronic Frontier Foundation questions the legality of the kiosks because they build patron databases that California law forbids. The state's ID privacy statute bars businesses from retaining or using scanned ID data except to verify age or prevent fraud; Joe Mullin wonders why such a serious violation of this law goes unchecked. Californians should be deeply concerned about businesses that collect information from government-issued IDs and use it to build databases about where people go, whom they associate with, and whether they should be allowed into other public gathering places. That concern is
Aug 1, 2026 · via boingboing.net
â Massachusetts Institute of Technology ( MIT ) is reportedly spending more than $3 million to install over 500 AI-powered surveillance cameras across its campus as part of a multi-phase security expansion. While the university says the project is intended to improve campus safety, students have raised concerns over facial recognition capabilities, expanded surveillance and how the collected data could be used. â According to MIT's student newspaper, The Tech, these new cameras will be capable of real-time face and object classification, including detection of motion, crowds, face masks, and other attributes. Defending the rollout, MIT spokesperson Kimberly Allen said the cameras are "part of regular efforts to promote campus security," adding that the data collected is "retained up to 30 days" unless an exception is granted. â More than 500 AI cameras planned across campus â According to documents reviewed by The Tech, installation of the new surveillance system began in November 2025 and is expected to continue through September 2026. â The project includes 521 AI-integrated cameras across academic buildings, residence halls and outdoor locations along Memorial Drive. Of these, 501 are single-lens cameras, and 20 are quad-lens models capable of monitoring multiple directions simultaneously. â The expansion will also add 67 exterior cameras and 13 new "Code Blue" emergency phone towers, each equipped with surveillance cameras, public address systems, emergency lighting and pan-tilt-zoom cameras. â Most of the project's nearly $2 million camera contract has been awarded to Siemens, while telecommunications contractors LCN and Picardi Electric are handling installation, wiring and related infrastructure work. â Cameras use AI for real-time detection â The primary camera supplier is Hanwha Vision, whose Wisenet AI cameras use deep learning algorithms to identify and classify people, vehicles, and other objects in real time. â According to technical specifications cited by The
Aug 1, 2026 · via timesofindia.indiatimes.com
Navega por la fotografía y haz zoom. E identifica a cada olímpico clicando sobre las caras.
UNA FOTO PARA LA HISTORIA
Aug 1, 2026 · via as.com
A dedicated van and officers were there this week in an effort to target shoplifting, anti-social behaviour and drug-related crime.
LFR captures real-time video feeds of people’s faces, comparing them instantly against a pre-determined police watchlist to generate alerts.
The method is controversial, but police say anyone not on its watchlist has their image deleted straight away.
Aug 1, 2026 · via radiojackie.com
The best facial recognition AI tool depends entirely on what you are trying to do. A journalist verifying a source’s identity needs a different tool than a developer building KYC into a fintech app, and both need different tools than a security team deploying access control across a 50-camera facility. The global facial recognition market is projected to reach $14.55 billion by 2031, growing at 16.79% annually. But most “best facial recognition tools” articles dump consumer search engines, developer APIs, and enterprise platforms into one undifferentiated list, which helps nobody make a decision. This article splits the category into three sections matching the three buyer profiles: consumer face search tools (for finding where faces appear online), developer APIs (for building facial recognition into your product), and enterprise security platforms (for physical access control and surveillance). Pick the section that matches your use case. Part 1: Consumer AI Face Search Tools These tools let individuals upload a photo and find where that face appears across the internet. The primary users are people verifying online identities, checking dating profiles, monitoring their own image, and conducting OSINT investigations. 1. FaceCheck ID: Best for Scam Detection and Social Media Search FaceCheck ID searches social media profiles, dating apps, mugshot databases, sex offender registries, and scam report sites. Its defining feature is the safety layer: when a match appears on a known scam profile or offender registry, the tool flags it with a red warning icon (Scam Alert, Sex Offender, Many Accounts) before you click through. FaceCheck ID handles low-quality images (blurry screenshots, side profiles, cropped photos) better than any consumer face search tool I have tested. I ran the same blurry screenshot from a WhatsApp video call through both FaceCheck ID and PimEyes. FaceCheck ID returned 4 usable matches within 8 seconds. PimEyes rejected
Aug 1, 2026 · via autogpt.net
Why Veridas and Fourthline are building a new kind of identity platform Consolidation in the biometrics industry continues, as firms find synergies in technology, regulatory coverage, and customer base. Recently, Veridas announced a merger with Fourthline. Eduardo Azanza, CEO of Veridas, says the two companies complement one another in terms of geography, the type of customers they’re courting, and their product portfolio. “Some people, they might say, ‘they do more of the same’,” Azanza says, on the latest episode of the Biometric Update Podcast. “That’s not the case. As long as they’re delivering their systems for NEOs, NEO banks, NEO cryptos, etc., they’ve deployed a full end-to-end platform to do the KYC. As long as we’re working with more traditional banks, telcos, etc., we’ll be more vertical on producing biometric engines, anti-fraud capabilities, also the wallet capability that complements nicely with their KYC, payment sanction, anti-money laundering services and bank account verification.” “So, let’s say that the overlap in terms of technologies is minimal.” Azanza says identity remains fairly fragmented, but large clients increasingly want hassle-free products that cover the gamut of services – and divergent regulatory regimes demand flexibility. “The idea is that you can orchestrate with the very same components very complex, different journeys for different regulations,” he says. But a critical piece is owning the technology: “if you do not own the tech, as long as the challenges evolve on a monthly basis, you cannot keep up the pace.” For more on identity verification, the evolution of fraud and some novel insights on why digital ID is an environmentally friendly choice, check out the full episode. Listen to this episode at: Spotify, Apple, YouTube, Podbean Runtime: 00:26:32 Article Topics acquisitions | Biometric Update Podcast | biometrics | digital identity | Fourthline | Veridas Comments
Aug 1, 2026 · via biometricupdate.com
5 Apps That Let Your Phone's Camera Identify Almost Anything Gone are the days when smartphones used to be more about themselves and less about the cameras. Today's phones are built around cameras and not the other way around. Your phone's camera is no longer just a way to take photos and videos; it's basically a pocket encyclopedia that can tell you about anything you point it at. Be it a mystery plant that you spotted on the road or a pair of sneakers that you found your friend wearing. This shift did not happen overnight. It's the result of years of AI and image-recognition work quietly baked into apps most people already have installed. Your smartphone's camera has turned into a translator, a plant guide, a personal shopper, and even a stargazing companion. The best part is that for most tasks, you don't need to install any third-party apps. You can use Google Translate or iPhone's built-in translator to convert text from one language to another and have it read aloud. No matter what app you're using, it has the same underlying idea — your camera is doing the identifying so you don't have to guess or do a web search manually. We have compiled a list of the best apps that can let you learn about or understand things just by pointing your camera at them. Each one lets you learn about different categories like plants, fashion, home decor, and general objects just by pointing your camera at them, so you're never stuck wondering. Google Lens Google Lens is the closest thing you can get to an app that identifies anything your camera can see and tells you about it. When you start using Google Lens, you'll understand (and might appreciate) how far technology has advanced. Built into
Aug 1, 2026 · via slashgear.com
You need to enable JavaScript to run this app.
Aug 1, 2026 · via qualcomm.com
Google Deepmind unveils Gemini Robotics 2 to power robots of all shapes from tabletop arms to humanoids Google Deepmind has introduced Gemini Robotics 2, which it calls its most advanced vision-language-action (VLA) model yet. VLA models combine image recognition, language processing, and action control to help robots operate in physical environments. Deepmind says the model can control systems ranging from tabletop arms to full-body humanoid robots. The company describes Gemini Robotics 2 as an "intelligence layer" for a new generation of adaptive robots. It can manage full-body movement, perform fine motor tasks, and coordinate multiple robots, according to Deepmind. Developers can apply for early access through the waitlist. Google Deepmind also introduced Gemini Robotics ER 2, a model designed for "embodied reasoning." The term refers to understanding the physical world and deciding which actions to take based on that information. ER 2 acts as a higher-level control system for robots and replaces Gemini Robotics ER 1.6, released in April. The new model is available in Google AI Studio. AI News Without the Hype – Curated by Humans Subscribe to THE DECODER for ad-free reading, a weekly AI newsletter, our exclusive "AI Radar" frontier report six times a year, full archive access, and access to our comment section. Subscribe now
Aug 1, 2026 · via the-decoder.com
Huedoku is a daily color puzzle with simple rules and a satisfying solve. It's like sudoku, but with colors instead of numbers.
Sign up here to get notified every time we publish a new Huedoku!
Have a great weekend — Huedoku #79 drops Monday! 🌈
🌈 New Huedoku drops every weekday at 4:00 a.m. PT / 7:00 a.m. ET.
Comments
Jul 31, 2026 · via buzzfeed.com
Italy's proposed policy to store facial biometric data for seven days from individuals captured on video in sensitive areas is facing scrutiny over its compliance with European Union laws, as reported by Biometric Update.The Italian Senate's EU Policies Committee approved a decree allowing the storage of facial biometric data from cameras in sensitive public areas for seven days. This move has drawn criticism from the data protection authority, Autorità Garante per la Protezione dei Dati Personali (GPDP), which deems the policy inconsistent with the EU's AI Act restrictions on live facial recognition. The GPDP argues that continuous biometric data collection and retrospective processing for investigations could be interpreted as live facial recognition, which is generally prohibited under the AI Act, except for specific exceptions like searches for missing persons, imminent terrorist threats, or identification of serious offenses.A representative for Premier Giorgia Meloni's office stated that Italy remains compliant with European regulations, including the AI Act. The debate intensified following recent public order incidents, including riots and violent protests. The European Commission has emphasized that facial recognition in publicly accessible spaces is prohibited but noted that a definitive assessment of the Italian policy requires a detailed review.Source: Biometric Update Get daily email updates SC Media's daily must-read of the most current and pressing daily news You can skip this ad in 5 seconds
Jul 31, 2026 · via scworld.com