Zero-Shot Local Document Parsing with Gemma 4: Treating PDFs as Images Treating PDFs as images and feeding those images to Gemma 4 dissolves the scanned-versus-digital distinction that makes every text-extraction pipeline fragile. Fix that. # Introduction Run pdfplumber on a scanned invoice, and you get nothing. Run it on a multi-column research paper, and you get a stream of text that has lost every spatial relationship the layout encoded. Run it on a filled PDF form, and you get the field labels concatenated with the values in reading order, with no way to tell which belongs to which. Text-extraction tools have one assumption baked in: the PDF has a selectable text layer. The moment that assumption fails — scanned documents, image-only PDFs, complex form layouts, anything with merged table cells — the tools fail silently. You get empty output or garbled text, and the failure mode gives you no signal about what went wrong. The image approach sidesteps this entirely. Render each PDF page to a high-resolution image. Feed that image to a vision-language model. Ask it what you need in plain language. No optical character recognition (OCR) pipeline, no layout parser, no template matching per document type. The model reads the page the way a human reads a printed page. Gemma 4, released by Google DeepMind on April 2, 2026, with a full Apache 2.0 license, lists Document/PDF parsing as an explicit capability alongside OCR, chart comprehension, handwriting recognition, and screen understanding. It runs entirely locally. No API key, no cloud call, no data leaving your server. The project thread through this article is a local document intake pipeline that processes supplier invoices, extracting vendor name, invoice number, line items, totals, and due date, and outputs structured JSON. It works on scanned and digital PDFs alike. # Why Treat
Jul 7, 2026 · via kdnuggets.com
Big Tech has spent the last 12 months making steady inroads into fashion, and Vogue Business has been clocking every move. So it came as little surprise when last week Meta revealed a Kylie Jenner-fronted campaign for its latest AI smart glasses line, a wider range of 26 new Meta Glasses styles and one pair co-designed with Jenner herself. Like its existing Ray-Ban Meta and Oakley smart glasses lines, the new designs are developed with EssilorLuxxottica — the group that also owns Prada eyewear, rumored to be Meta’s next collaborator. Snap released its second attempt at smart glasses, the new $2,195 AI and AR-powered Specs glasses, just a week earlier, and Google unveiled the first designs of its upcoming Intelligent Eyewear AI smart glasses a fortnight before that, so it was high time the spotlight swung back to smart glasses incumbent Meta, which first launched its version back in 2021. Big Tech has been borrowing from fashion’s playbook to improve its image, and the new Kylie partnership and glasses line launch is its biggest pivot to fashion and culture yet. Beneath the question of whether Jenner can finally make Meta’s smart glasses cool, there’s a much more challenging question for Meta and its rivals. Is recruiting fashion’s most influential tastemakers enough to normalize a product that fundamentally makes consumers so uneasy? Not so fast. Initial reactions to the Jenner drop focused on one specific element of the glasses: their built-in camera. “Just another way for Meta to spy on you by seeing what you see everyday,” said one Instagram user. “The people don’t want this.” “These should be illegal,” said others. “No surveillance state,” said one more. Similar comments flooded in, in response to Snap’s Specs and Google’s Gentle Monster collaborations. “Booooo we hate surveillance technology disguised as fashion,” one
Jul 7, 2026 · via vogue.com
Hamad International Airport rolls out biometric travel at over 700 touchpoints Passengers departing from Hamad International Airport can now complete nearly every stage of their journey using only facial recognition after the airport introduced one of the world’s largest biometric passenger-processing systems. Developed in partnership with Qatar Airways and aviation technology provider SITA, the new Fast Pass service connects more than 700 biometric touchpoints across the airport. It allows eligible passengers to check in, drop bags, clear security and board their flight without presenting a passport or boarding pass at each step. Passengers can enrol in Fast Pass either through the Qatar Airways mobile app during check-in or at a self-service kiosk in the terminal. The enrollment process takes only a few seconds, after which facial recognition becomes their primary form of identification throughout the departure journey. One biometric identity from check-in to boarding The rollout represents a significant expansion of biometric passenger processing, with facial verification integrated into self-service bag-drop units, security checkpoints and boarding gates. The SITA system prevents repeated document checks while maintaining high standards of security and data protection by verifying passengers once before their journey and by automatically identifying them at subsequent touchpoints. Passengers who don’t wish to participate can opt for traditional processing. The airport plans to expand the service to include Qatar Airways transfer passengers and, eventually, other airlines operating from Hamad International Airport, helping streamline connections for transit travellers. Meeting growing demand for contactless travel The launch comes as airports worldwide seek to accommodate growing passenger numbers while minimising congestion at key processing points. SITA cited research from the International Air Transport Association (IATA) showing that most passengers now prefer biometric identification over physical travel documents. IATA published the results of its Global Passenger Survey last year, showing that: - 85% of
Jul 7, 2026 · via aerospaceglobalnews.com
BuzzFeed GamesOnly People With Elite Pattern Recognition Can Solve This Color Puzzle Called HuedokuHuedoku #59! New week, fresh puzzle — let’s kick things off with some color. 🌈Posted 6 hours agocommentFacebookPinterestLinkby Crystal RoBuzzFeedBuzzFeed StaffHi, I’m Crystal, a Senior Editor based in Los Angeles and creator of BuzzFeed’s “That Got Dark” newsletter. 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! Come back tomorrow for Huedoku #60 — and share your score to challenge a friend! 🌈 🌈 New Huedoku drops every weekday at 4:00 a.m. PT / 7:00 a.m. ET. 🧩🗂️ Huedoku Archive — Every past puzzle, all in one place! Comments
Comments
Jul 7, 2026 · via buzzfeed.com
Sainsbury's, the UK's second-largest supermarket, is significantly increasing its use of facial recognition technology across its stores to deter shoplifters. The move, which will see the technology deployed in up to 200 locations by the end of 2026, has drawn criticism from privacy advocates who deem it a violation of privacy rights, as reported by The Register.The supermarket chain is tripling the number of stores utilizing facial recognition, expanding from over 55 current locations to a projected 200 by year-end. Sainsbury's claims the system, provided by Facewatch, has been effective, with 90% of identified individuals not returning to the stores. This expansion follows trials that began last year. Privacy campaigners, including Big Brother Watch, have labeled the deployment "shameful" and a serious threat to privacy, urging shoppers to boycott the supermarket. Concerns have been amplified by incidents such as a shopper being wrongly ejected from a store due to a facial recognition alert, highlighting potential inaccuracies and the impact on innocent customers.Despite apologies and promises of staff training following such errors, critics argue that mass surveillance is not a justifiable response to shoplifting and that innocent shoppers should not be subjected to such identity checks.Source: The Register 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 7, 2026 · via scworld.com
We have a habit of dismissing social media trends as inane and vapid while ignoring the disturbing undercurrent. A few weeks ago I was reminded of that when I saw an Instagram carousel by British fashion personality Alexa Chung. Shared with her 6 million followers, she showed different outfits through screenshots of herself entering and leaving her home on her security camera. Rita Ora commented, “Good angle keep this series going”. Security system company Ring commented, “Fit checks on Ring cam? Next level.” The post caught my eye among the feed of curated noise, a counterculture take on the traditional iPhone outfit photo. Its presumed effortlessness felt intimate and off the cuff. Social media loves that sort of thing. But something about it didn’t sit right with me. The fish-eyed lens and zoom-in and zoom-out icons made me feel voyeuristic, like I was stumbling on private footage I shouldn’t be seeing. Maybe that’s exactly what it was. It called to mind Ring’s dystopian Super Bowl ad from earlier this year. The story of finding a lost dog using neighbours’ Ring cameras and AI tools was intended to pull at heartstrings. Instead, it was a revealing confession of how the Amazon-owned tech company leeches itself on to communities. Its ability to use facial recognition software, not to mention how it partners with local law enforcement, is deeply concerning. On TV, we’re sold heartwarming stories about pets found via private cameras. On social media, we see influencers turn security footage into “fit checks”. In our suburbs, we face hyper-surveillance in our grocery store chains. At the same time, there’s been a rise in in-home CCTV social media content and people modifying old CCTV cameras for personal use. We now don’t blink an eye at strangers being filmed in public. Taken individually, we
Jul 7, 2026 · via theguardian.com
Explore the things
you love
.
Log into Facebook
Email or mobile number
Password
Log in
Forgot password?
Create new account
English (US)
Español
Français (France)
中文(简体)
العربية
Português (Brasil)
Italiano
More languages…
Sign Up
Log In
Messenger
Facebook Lite
Video
Meta Pay
Meta Store
Meta Quest
Ray-Ban Meta
Meta AI
Instagram
Threads
Privacy Policy
Consumer Health Privacy
Privacy Center
About
Create ad
Create Page
Developers
Careers
Cookies
Ad choices
Terms
Help
Contact Uploading & Non-Users
Meta © 2026
Jul 7, 2026 · via facebook.com
SIA Opens Applications for Four Youth Recognition Programs The Security Industry Association is offering awards and educational scholarships to support emerging professionals and students. - By Jesse Jacobs - Jul 06, 2026 The Security Industry Association has opened the application and nomination process for four of its premier workforce development initiatives aimed at emerging security professionals and students. The 2026 programs include the annual 25 On the RISE awards alongside three specialized educational scholarships. According to association leadership, the initiatives are designed to provide financial assistance and networking access to help younger industry professionals further their career goals. The 25 On the RISE award program, organized by the association’s RISE community, honors 25 professionals under the age of 40 or with fewer than two years of industry experience. Winners are recognized for advancements in diversity, innovation and corporate impact, and receive funding to attend the Securing New Ground conference in New York City. The Denis R. Hébert Identity Management Scholarship, named after the former SIA board chair, provides two $5,000 scholarships to young professionals at member companies. The funds must be applied toward post-secondary education, certifications or training programs specifically within the identity management sector. The James Rothstein Business Scholarship, named for another past SIA chair, funds conference passes and travel expenses for students and RISE community members to attend the executive-level Securing New Ground conference. The SIA–SecuritySpecifiers Young Consultants Scholarship supports security consultants, designers and specifiers aged 35 or younger. Recipients receive full admission and travel accommodation to the CONSULT symposium in Kansas City, Missouri, where they participate in dedicated mentorship meetings with architectural and engineering program sponsors. Association officials stated that the combined programs are part of a broader, long-term industry mission to cultivate and retain next-generation leadership talent. The deadline to submit applications and nominations for all
Jul 6, 2026 · via securitytoday.com
TSA expands biometric identity checks to airline crews Airline pilots, flight attendants, and other eligible crew members are beginning to encounter a materially different way of entering airport sterile areas as the Transportation Security Administration (TSA) replaces Known Crewmember (KCM) with a new facial comparison program called Crewmember Access Point (CMAP). The change is more than a technology upgrade. Known Crewmember was created in 2011 as an industry operated system that allowed vetted crewmembers to present an airline ID and KCM barcode at designated access points. Until now it had been a joint initiative between the Air Line Pilots Association and Airlines for America. Under CMAP, TSA will assume program administration, use data already supplied by carriers through federal crew listing programs, and photograph crew members at access points for comparison against an image held in federal databases. TSA says the transition began June 22 and is scheduled to conclude at airport locations by September 30, although KCM is not expected to formally sunset until the end of this year. The first locations were Washington Reagan, Washington Dulles, and Las Vegas. A second wave began the week of July 5 at airports including San Diego, Honolulu, Salt Lake City, Austin, and Phoenix, with 30 additional airports scheduled for later July conversion dates. Those dates remain subject to change. CMAP remains voluntary in a formal sense. Crew members must affirmatively consent before their airlines add them to the program, and those who decline may use regular passenger screening to reach the sterile area. For crew members, the practical choice is between consenting to facial comparison or losing access to the separate expedited lane that KCM provided. A failed biometric match or a random selection for additional screening sends the crewmember to a passenger checkpoint. The program relies on data that airlines
Jul 6, 2026 · via biometricupdate.com
INTELLIGENT psoriasis research is accelerating precision diagnosis, personalized treatment, and long-term management across dermatology. Intelligent Psoriasis Research Enters a New Era A bibliometric analysis of intelligent psoriasis research from 2005 to 2025 highlights a growing focus on digital tools that could support diagnosis, treatment selection, and disease monitoring. The study examined literature related to intelligent diagnosis and treatment of psoriasis in the Web of Science Core Collection, using VOSviewer and CiteSpace to assess co-cited literature, keywords, research trends, and patterns of international collaboration. Psoriasis remains a chronic inflammatory disease driven by genetic, immune, and environmental interactions. The central mechanism described in the analysis is aberrant activation of the Th17/IL-23 pathway, contributing to uncontrolled keratinocyte proliferation and systemic inflammation. The condition affects an estimated 1–3% of the global population and carries substantial clinical and quality of life burdens. Digital Tools Could Advance Precision Dermatology The analysis positions intelligent diagnosis and treatment as a potential pathway toward more individualized psoriasis care. Intelligent image recognition and multimodal data fusion were highlighted as key areas that may help clinicians integrate visual, clinical, and other patient level data to improve diagnostic accuracy and guide treatment decisions. These approaches could be particularly relevant in a disease where symptoms extend beyond visible plaques. The study notes that 70–90% of patients experience moderate to severe pruritus, 50% experience nail thickening or loss, and up to 30% have associated arthropathy. Psychological burden is also prominent, with 59.1% of patients affected by social discrimination, further reducing quality of life. Broader Applications Beyond Psoriasis By mapping research activity in intelligent psoriasis research, the study identifies a developing field with potential implications for precision medicine and long term disease management. The authors suggest that intelligent diagnostic and therapeutic systems may help reduce medical costs, improve long term management, and support more personalized
Jul 6, 2026 · via emjreviews.com
DC Superior Court Judge Neal Kravitz ordered prosecutors to provide more information on an artificial intelligence (AI) facial recognition software program that was used to identify a shooting defendant on July 1. Marquis Foster, 43, is charged with assault with intent to kill while armed, aggravated assault while armed, and two counts of possession of a firearm during a crime of violence or dangerous crime for his alleged involvement in a non-fatal shooting that wounded one individual on June 8, on the 2500 block of Georgia Avenue, NW. The individual sustained gunshot wounds to his chest, arm, and hip. Foster was identified using surveillance footage of a suspect taken from nearby the crime scene. Police then reportedly used a facial recognition program, Clearview AI, to scan over 30 billion publicly available facial images for comparison. The image of the potential suspect that was identified by the software program was then imported into a Metropolitan Police Department (MPD) mugshot’s database, where police identified Foster from an existing image. The victim identified Foster based on a photo line up that pictured eight other individuals. He said he did not recognize any of the other photographs. During the hearing, Foster’s lawyer, Elizabeth Weller, said “I’ve never heard of this before,” and requested more information about the process. The prosecution said they are trying to get more information from the police. Judge Kravitz told the prosecution that Weller is entitled to information about the program and he expects them to provide it. Parties are scheduled to reconvene on July 22.
Jul 6, 2026 · via dcwitness.org
Artificial intelligence did not suddenly appear with ChatGPT. It evolved over four decades through advances in computer graphics, game development, machine vision, and parallel computing. Graphics processors, originally built to draw pixels, gradually became programmable computing engines capable of training neural networks and running AI models. JPR followed that evolution from the beginning, covering graphics hardware, GPU computing, machine learning, and AI processors as each technology emerged. Looking back, the progression seems remarkably logical, even if it rarely felt that way at the time. I might not exist but for you. People often ask when JPR began covering artificial intelligence. The answer depends on what they mean by AI. Today’s AI refers to large language models, generative AI, autonomous agents, and foundation models. Forty years ago, AI meant something entirely different. Researchers talked about expert systems, neural networks, fuzzy logic, machine vision, and pattern recognition. Game developers used the term to describe scripted routines that controlled non-player characters, while engineers applied AI techniques to CAD, robotics, and image analysis. JPR and its predecessor, JPA, covered many of those technologies years before anyone imagined conversational AI. When Jon Peddie Associates opened its doors in 1985, the graphics industry stood at the beginning of its transition from fixed-function hardware to programmable computing. Graphics workstations powered CAD, scientific visualization, and digital content creation. Researchers already experimented with neural networks and knowledge-based systems, although computing power limited what they could accomplish. AI remained an enabling technology rather than a market of its own. The first AI many consumers encountered appeared in games. During the mid-1990s, developers programmed enemy behavior with lookup tables, decision trees, and finite-state machines. Characters reacted to player actions, navigated environments, and coordinated attacks through carefully designed logic rather than learning. In 1994, Matrox demonstrated Sentõ, a 3D game that showcased
Jul 6, 2026 · via jonpeddie.com
Artificial Intelligence Automatically redact PII in images with Amazon Nova Sharing data internally across teams, externally with partners, or using it for workloads such as machine learning (ML) model training is fundamental to modern business operations. However, when that data contains Personally Identifiable Information (PII), organizations face significant legal and compliance obligations under regulations such as the General Data Protection Regulation (GDPR) and the Payment Card Industry Data Security Standard (PCI DSS). If PII isn’t properly redacted before sharing or processing data, the result can be regulatory penalties, reputational damage, and erosion of customer trust. PII redaction in real-world image datasets is particularly challenging. Unlike structured text, PII in images can appear in unexpected places and forms: a partial face captured at the edge of a frame, a face reflected on the polished surface of a car, a partially visible street sign that, combined with other visual cues, becomes identifiable, or a document lying on a desk in a wide-angle photo that reveals names, addresses, or ID numbers. These edge cases routinely defeat single-purpose masking tools. Amazon Nova is a family of foundation models with advanced vision understanding capabilities, making it a strong candidate to serve as the intelligent coordinator for complex image analysis workflows. Nova interprets image content holistically, reasons about whether something constitutes PII in context, including the subtle and unusual cases described earlier, and directs the entire redaction pipeline from start to finish. By understanding the “what” of PII, Nova coordinates specialized tools to achieve pixel-level precision in redaction while preserving the overall value of the image. In this post, we present a multi-step pipeline directed by Amazon Nova, which uses its contextual vision reasoning to coordinate complementary tools, including Meta’s open-source Segment Anything Model (SAM 3) deployed on Amazon SageMaker AI for pixel-level segmentation, and Amazon
Jul 6, 2026 · via aws.amazon.com
For travelers enrolled in TSA PreCheck, getting through airport security without ever pulling out an ID is becoming a little more straightforward. The Transportation Security Administration (TSA) has partnered with Google Wallet, Google’s digital wallet app that can be used for storing credit cards, boarding passes, and digital ID on Android devices, to simplify enrollment in its growing Touchless ID program. The program uses facial recognition to quickly verify the identity of TSA PreCheck members at airport security checkpoints instead of requiring them to hand over a driver’s license or passport. The update doesn’t change what happens at the checkpoint, but it does streamline one of the more confusing parts of using Touchless ID. Until now, travelers generally had to enroll through each participating airline by uploading their passport information to a frequent flier account or airline app. If you flew multiple carriers, you often had to repeat the process. With the new integration, eligible travelers can instead use a passport-based digital ID stored in Google Wallet as a single credential across more than 100 airlines participating in the TSA PreCheck Touchless ID program. The move reflects TSA’s broader push toward a more digital airport experience. Over the past several years, the agency has expanded support for mobile driver’s licenses, biometric screening, and digital identity verification. How TSA Touchless ID works For those unfamiliar with Touchless ID, the name is fairly literal. At participating TSA PreCheck security checkpoints, travelers enter a dedicated line for Touchless ID where a camera compares a live image of their face with the passport information they’ve previously shared. If the images match, they are cleared to continue through security without presenting a physical ID. Participation is voluntary, and travelers can opt for a traditional ID check instead. Touchless ID is now available at more than
Jul 6, 2026 · via afar.com
Home Secretary gives evidence on the work of the Home Office
6 July 2026
On Tuesday 7 July 2026 the House of Lords is hearing evidence from the Rt Hon Shabana Mahmood MP, Home Secretary.
Purpose of the session
This is a one-off session examining the work of the Home Office. The Committee are expected to explore the Immigration and Asylum Bill announced last week, the Common Travel Area, and police use of facial recognition technology.
Possible themes include:
-
The Immigration and Asylum Bill
-
The Entry/Exit System
-
Facial recognition technology
-
An update on the Fairer Pathway to Settlement proposals
-
Progress on the Safer Streets mission
Further information
Jul 6, 2026 · via committees.parliament.uk
Abstract Machine vision and AI-based defect detection systems are increasingly deployed in manufacturing to support consistent product quality and high production efficiency. However, these automated inspection systems often suffer from sensitivity to imaging variability, dependence on large labeled datasets, and the need for manually engineered preprocessing pipelines–limitations that hinder accuracy and reliability in real industrial conditions. This study presents a novel approach for optimizing image preprocessing strategies for neural network–based defect detection using a genetic algorithm (GA)-driven evolutionary framework. The method systematically explores a set of 48 preprocessing operations and automatically evolves optimal filter sequences through multi-objective fitness evaluations incorporating classification accuracy, computational efficiency, and preprocessing robustness. The genetic algorithm generates diverse preprocessing sequences of varying lengths (3–6 filters) and evaluates a broad range of population sizes (20–100 individuals) and generation limits (20–150 generations) to identify configurations that maximize detection performance while reducing data requirements. Extensive experiments across three product categories show that GA-optimized preprocessing significantly outperforms raw-image baselines and manually designed preprocessing pipelines. Results demonstrate substantial gains in classification accuracy (up to 15%) and improved data efficiency, requiring 30–60% fewer training images to achieve target performance. The findings confirm that evolutionary optimization provides a robust and scalable solution for industrial defect detection, enabling more reliable and efficient machine vision systems for modern manufacturing environments. Similar content being viewed by others Introduction Machine vision integrated with artificial intelligence (AI)-based defect detection is widely utilized in manufacturing and production environments to ensure product quality, regulatory compliance, and operational efficiency. These automated inspection systems play an instrumental key role in detecting, identifying, and localizing defects, thus maintaining consistency and reliability throughout the production cycle. As demand for precision and automation continues to grow, improving the accuracy and robustness of defect detection systems is essential to reduce variability in final products and maximize
Jul 6, 2026 · via nature.com
Abstract Deep neural networks have achieved remarkable success in image recognition tasks, yet they remain vulnerable to carefully designed input perturbations that can cause incorrect predictions while producing little visible change in the original image. In this study, we introduce the Turing Deimatic Attack (TDA), a biologically inspired adversarial attack that generates structured perturbations using reaction-diffusion processes that mimic natural pattern formation. Unlike conventional approaches that rely on gradient information from the target model, TDA operates without model queries and creates coherent spatial patterns from a compact set of control parameters. We evaluated TDA on seven benchmark datasets spanning natural images, facial recognition, medical imaging, and traffic sign classification using both convolutional and transformer-based neural network architectures. The proposed method consistently reduced classification performance across all datasets while maintaining high visual similarity to the original images. Mean attack success rates reached 48.3% on Fashion-MNIST and 63.3% on Labeled Faces in the Wild, with individual models exhibiting success rates of up to 80.5%. Despite these performance reductions, image quality remained largely preserved, with structural similarity values exceeding 0.93 and perceptual similarity scores remaining below 0.10 across all benchmarks. Our experiments further reveal that model susceptibility varies with the interaction between the spatial structure of the perturbation and the features used by different architectures. Convolutional networks were generally more vulnerable on lower-resolution images, whereas transformer-based models became increasingly susceptible at higher resolutions. Ablation analyses indicate that attack effectiveness is associated primarily with the spatial organization of the generated patterns rather than with perturbation magnitude alone. These findings demonstrate that biologically inspired pattern-generation mechanisms can expose systematic weaknesses in modern vision systems and provide a practical framework for evaluating model robustness under realistic, spatially structured perturbations. Acknowledgements The author gratefully acknowledges the developers of the publicly available datasets and pretrained models used in
Jul 6, 2026 · via nature.com
Physical artificial intelligence (physical AI) technologies that understand the interaction between light and matter, perceive space, predict future situations, and act accordingly have been developed. They are expected to serve as a reference for implementing next-generation autonomous systems that operate in the real world, such as self-driving vehicles and humanoid robots. KAIST announced on the 6th that Professor Yoon Sung-ui’s research team in the School of Computing has developed four technologies: one that recognizes transparent objects such as glass and water, one that analyzes the interaction of light and matter to understand surrounding environments, one that enables robots to navigate to a destination using a single photograph, and one that predicts future situations to plan actions. These achievements were reported in four papers. Two were presented as oral talks and two as highlight papers at the International Conference on Learning Representations (ICLR 2026) and the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR 2026). The team developed a visual technology for recognizing transparent environments called ‘GLINT’, enabling AI to accurately perceive transparent objects such as glass. Conventional AI systems struggle to properly separate objects reflected in glass from the scenery beyond it. GLINT separates and analyzes both the reflections on the glass and the objects behind the glass. They also developed ‘RadioGS’, a technology that understands light and material properties and reconstructs scenes. By enabling AI to grasp how light hits an object, reflects, and scatters, the system can accurately infer the material of objects and the surrounding environment even when the lighting conditions change. The team also created ‘Visual-RRT’, an image-based technology for planning robot paths, thereby finding a way to connect visual information to actual behavior. Conventional robots required coordinate data for the destination, but with Visual-RRT, the robot compares the scene it sees with a target
Jul 6, 2026 · via dongascience.com
5 Android Phone Apps You Can Safely Uninstall There are many reasons why a phone may come with unwanted software preinstalled. Perhaps you purchased a Samsung phone that comes with a suite of software from the company, which is how the company offers features only Samsung Galaxy devices have. Or maybe you got a deal at your carrier for the latest handset and it came with a bunch of carrier bloatware. Even if you buy a Pixel directly from Google, these devices can offer their own bloat with extraneous apps like Google TV. No matter which carrier or manufacturer you prefer, there are a handful of preinstalled apps that came with your phone that can be safely uninstalled, or at the very least, hidden. If you've used a Samsung phone, there's a high chance it came with a Meta app or two preinstalled (some are even hidden) or a fleet of Microsoft Office apps you don't need. Worse, many OEMs duplicate Google's work, offering their own file browsers, image storage solutions, and note-taking apps. Samsung notoriously offers its own suite alongside Google's, but it's not alone either. Why would any manufacturer or carrier leave that potential revenue stream for Google to eat up when they can easily create their own proprietary version? All of this bloat exists to compete for your attention or data, and the truth is, most is unnecessary. So rather than live your Android life with bloatware, we're here to share the more prominent apps you can safely uninstall, from unnecessary carrier apps to redundant security apps. Here's your cheat sheet so you know what you can safely remove or hide in order to transform your Android experience into something much more easily navigated and purpose-driven. It's time to take control of your Android phone. Carrier apps
Jul 6, 2026 · via bgr.com
Warakorn Luangluewut, Researcher, Defense Technology Institute (DTI), Thailand By Sol Gonzalez Meet the young public sector officials in the inaugural Young & Official Report 2026. Warakorn Luangluewut, Researcher, Defense Technology Institute (DTI), Thailand. Image: Warakorn Luangluewut. 1) What does public service mean to you? Can you share more about your role in the public sector? I am primarily engaged in research related to defense technology, with responsibilities encompassing research activities, the development of Artificial Intelligence (AI) applications, and communication system development. My professional background began with studies and practical work in image processing, signals and systems, machine learning, and other related fields. I have subsequently applied this knowledge to AI, communication technologies, and other relevant areas in order to maximise benefits for myself, the public, and the nation. In my work, I strive to develop expertise and innovations in areas of personal interest while ensuring alignment with national needs and organisational policies, so that the outcomes can deliver the greatest possible benefit to all stakeholders. 2) Tell us about a project you championed. What impact did it have on the community? My work provides benefits across the military, civilian, and research sectors. In the military domain, one of my recent projects involves the development of a system for detecting whether training postures are performed correctly. This system helps make training more convenient, faster, and more efficient by introducing an application that supports instructors in their work. As a result, instructors are able to supervise and evaluate trainees more effectively, especially in situations where the number of instructors is significantly smaller than the number of cadets. Therefore, this work plays an important role in enhancing military training efficiency. In addition, I have also worked on the development of a flood area detection system, which can be further extended into an automatic
Jul 6, 2026 · via govinsider.asia