Following IPO, ROC is investing in homegrown security for US market In February, Colorado-based biometrics and vision AI provider ROC closed the first big biometrics IPO of 2026, raising just over $24 million at $6 per share. ROC CEO Scott Swann says it would have been faster to look to private markets for capital that could help it meet some of its objectives, as it expands the scope of its business . But the company “saw the IPO as infrastructure.” “We saw it as a way that we could scale in a really disciplined way. A big part of this was so that we could maintain our culture and our independence.” That same culture is part of what makes ROC appealing to shareholders. “I think that our growth story resonated with people as we described what we were trying to achieve,” Swann says. ROC has always positioned itself front and center as an American-made company. As global political and economic relationships take new shapes, it stands positioned to provide homegrown defense and security capabilities to the U.S. market. “The timing is right,” Swann says. “I think people understand supply chain more now than they ever have. And as we think about artificial intelligence, I think we’ve seen examples where bias can be built into these models. So, you have to have a lot of trust in the entity that builds these models out for you.” “And so, for us understanding the whole geopolitical environment today and being able to really be at the forefront of putting the United States as a leader in this position, that’s something we’re completely committed to doing.” Moving further afield from its origin as a components provider, ROC also recently launched biometric physical access control software, reflecting significant growth forecasted in the biometric physical access
Apr 11, 2026 · via biometricupdate.com
Abstract Concrete structures are a vital component of urban infrastructure, requiring regular maintenance to ensure public safety and structural integrity. A crucial element of this maintenance is the identification of surface cracks, which have traditionally relied on manual inspection methods that were frequently work-intensive, subjective, and sometimes dangerous. This work presents a hybrid methodology that integrates deep feature extraction with machine learning classification for identifying structural deterioration in concrete components. A publicly accessible dataset comprising photos of both cracked and uncracked concrete surfaces was used. Deep features were extracted using VGG16, a convolutional neural network widely recognized for its success in visual pattern recognition. Several machine learning algorithms were used for classification of these features, including Artificial Neural Network, Decision Tree, Random Forest, Support Vector Machines and k-Nearest Neighbors. The experimental results indicate that, the highest accuracy was achieved by SVM (99.883%), followed closely by ANN (99.873%), k-NN (99.598%), and DT (99.580%), while RF performed the lowest (98.050%). Although not limited to seismic applications, the proposed method has the potential to be integrated into post-earthquake structural assessment workflows as part of structural health monitoring systems. Using deep learning and machine learning methodologies to detect damage in concrete infrastructure may enhance efficiency and precision, enhancing urban resilience and risk mitigation. Data availability The dataset can be reached from this link in the public repository “Mendeley data” website https://doi.org/10.17632/5y9wdsg2zt.2, under the title “Concrete Crack Images for Classification”. References Tang, S. W., Yao, Y., Andrade, C. & Li, Z. Recent durability studies on concrete structure. Cem. Concr. Res. 78, 143–154. https://doi.org/10.1016/j.cemconres.2015.05.021 (2015). Zar, A. et al. Towards vibration-based damage detection of civil engineering structures: Overview, challenges, and future prospects. Int. J. Mech. Mater. Des. 20 (3), 591–662. https://doi.org/10.1007/s10999-023-09692-3 (2024). Kabir, S. Imaging-based detection of AAR induced map-crack damage in concrete structure. NDT E
Apr 11, 2026 · via nature.com
SUMMARY This is AI generated summarization, which may have errors. For context, always refer to the full article. Albert, a 68-year-old retiree, received a call noon on August 14, 2025. The call came at a perfect time because Albert needed a payment reference number (PRN) from his SSS (social security service) but he was having a challenging time logging on to his app. The caller knew his full name, his SSS number, and his address, and offered to assist him in downloading a supposed updated SSS app. The assistance came in the form of him clicking a link provided over Viber, followed by a very long installation process. In one-and-a-half hours, his three bank accounts and two cash wallets — all connected to his Android phone — were swiped. More than one million pesos in life savings gone. Albert never provided any personal detail like OTPs (one time passwords) or security codes. “My dad was depressed for the first few months,” Albert’s daughter Jobelle Garcia told Rappler. The scam is a banking trojan deployed through malware on Android which can control your device from a remote location, and it has been traced to scam compounds in Cambodia, according to a first-of-its-kind report by US-based cybersecurity firm InfoBlox, and Vietnamese cyber safety non-profit Chong Lua Dao. “We uncovered an Android banking trojan that is likely operated from multiple locations, including the K99 Triumph City compound in Cambodia,” said the report. This is the first time that strong evidence has been found to connect a specific type of malware to a physical location of a scam compound, and it was made possible by trafficked workers who escaped from the compound and took damning evidence with them. The malware The team found a “sophisticated malware-as-a-service (MaaS),” which is a high-end malicious tool sold
Apr 11, 2026 · via rappler.com
The Federal Government has approved the deployment of a contactless biometric passenger verification system, known as VPASS, across Nigeria’s domestic airports as part of efforts to strengthen aviation security, enhance data integrity, and improve revenue generation. The Honourable Minister of Aviation and Aerospace Development, Festus Keyamo, disclosed this during the signing of the concession agreement today 9th April, 2026 at the Ministry’s Conference Room in Abuja. He explained that the initiative is designed to eliminate discrepancies in passenger data arising from inconsistent airline records, while also addressing unauthorized boarding practices. Keyamo said the system will ensure that all passengers on domestic flights are properly identified, thereby closing existing gaps that allow individuals to bypass Standard Identification Procedures. Keyamo noted that while strict identity verification measures are already in place for international travel, the VPASS system extends similar standards to domestic operations. ADVERTISEMENT He added that the system will subsequently be expanded to cover private aviation, to further strengthening security oversight across the sector. Keyamo described the project as a comprehensive reform, emphasizing that it will promote transparency, accountability, and safety in Nigeria’s aviation industry. He added that implementation will begin with infrastructure deployment by the concessionaire, followed by a nationwide sensitization campaign to ensure public awareness and compliance. Keyamo commended key stakeholders, particularly the Infrastructure Concession Regulatory Commission (ICRC) and the Federal Airports Authority of Nigeria (FAAN), for their roles in advancing the initiative. In his remarks, the Permanent Secretary, of the Ministry , Mahmud Adamu Kambari, FCNA, FICA, reaffirmed the Ministry’s commitment to modernizing the aviation sector through innovative, technology driven solutions aimed at improving operational efficiency, strengthening security, and enhancing passenger experience. ADVERTISEMENT Earlier, the Director of Commercial and Business Development at FAAN, Mrs. Adebola Agunbiade, described the VPASS system as a strategic step towards eliminating reliance on
Apr 10, 2026 · via tvcnews.tv
Dating App OkCupid: US Authority Criticizes Secret User Data Transfer Sensitive data from millions of users allegedly ended up with a biometrics startup. The FTC lets those responsible get away with mild conditions. Match Group is said to have shared extensive user data from its dating app OkCupid, including almost three million photos, with the biometrics company Clarifai. According to the US Federal Trade Commission (FTC), the photos were sent to the then-startup in 2014 along with location data and other personal information. There, they were used to train image recognition software. The operators are also said to have concealed the data transfer for almost twelve years and deliberately denied it to the public and concerned users. Possible consequences for those affected would be that their biometric data remains permanently in facial recognition systems and they become identifiable in other contexts. Furthermore, conclusions could be drawn about other intimate details, especially if the information is linked with other data sources or resold to other companies. Settlement without penalty The majority of OkCupid users are from the United States. The FTC documents do not specify which other countries may be affected. Match Group, which claims to be the world leader in online dating, is active in Germany primarily with apps such as Tinder or Hinge. Following a lawsuit by the FTC, the supervisory authority has now reached a settlement with the operating companies Match Group and Humor Rainbow. The operators do not admit the allegations of unlawful data transfer but commit to stricter data protection regulations, for which a fine is to be waived. Clarifai, in which the OkCupid founders were personally involved as investors, is said to have requested the data and received it without consideration and with no restrictions on its use. The affected users of the dating app
Apr 10, 2026 · via heise.de
Americans traveling to Europe are now encountering major changes at border crossings as the European Union’s long‑planned Entry/Exit System (EES) has officially come into force. The new system replaces traditional passport stamps with biometric checks, including fingerprints and facial scans, and is now fully operational across participating countries. The EES, which began limited rollout in late 2025, became fully active at airports and other external border points on April 10, 2026, according to the European Commission. Millions of U.S. travelers heading to Europe for vacations, business trips, cruises, and short stays are now subject to the updated procedures. What Has Changed As of April 10, border authorities across 29 Schengen Area countries—including France, Germany, Spain, Italy, Portugal, and Greece—are enforcing the new digital system. These countries, which allow passport‑free movement within their borders, now record all non‑EU entries and exits in a centralized database. The EES logs travelers’ biometric data and movement history, replacing the manual passport stamps used for decades. Who Must Use the New System The system applies to all non‑EU nationals entering participating countries for short stays of up to 90 days within any 180‑day period. This includes Americans traveling visa‑free for tourism, business, or family visits. EU officials say the system is designed to curb identity fraud, track overstays more effectively, and identify individuals who may pose security risks. Travelers entering the EU from the UK on routes where EU border checks occur on British soil—such as Eurostar services from London St Pancras, the Port of Dover, and the Eurotunnel terminal in Folkestone—now complete EES registration before departure. No biometric collection occurs in the United States. Passport Stamps Are Gone For U.S. travelers, one of the most noticeable changes is the elimination of passport stamping. Instead, border officers now collect: - Fingerprints - Facial scans -
Apr 10, 2026 · via newsweek.com
Trying to guess who someone is from their childhood photo is weirdly addictive. We’ve already done it with the world leaders, and it’s wild how much some of them still look exactly the same. That got me thinking: can you recognize former and current actors just from pictures of them as kids? Let’s see how good your guessing skills really are.
Comments
Apr 10, 2026 · via buzzfeed.com
Yamaha upgrades DXR/DXS and CXR/CXS mk3 series Yamaha upgrades DXR/DXS and CXR/CXS mk3 series The DXR mk3 powered loudspeakers and DXS mk3 powered subwoofers – along with their CXR/CXS mk3 unpowered counterparts – have been designed for live music, DJ sets, speeches and corporate events. The DXR mk3 range is available in three models: the DXR15, DXR12 and DXR10. Respectively featuring high durability 15-, 12- and 10-inch low-frequency drivers with 1.75-inch high-frequency units, all models are housed in a lightweight, injection-moulded enclosure. As well as weight reduction, this attenuates cabinet resonance, with optimised port positions to reduce standing wave effects. The DXS mk3 subwoofers share the same 96kHz DSP, power supply and rear LCD panel design, with a 2,500W Class-D amplifier and long-excursion 18-, 15- or 12- bass cones, plus 4-inch (DXS18 mk3) or 3-inch voice coil. The DXS18, DXS15 and DXS12 are housed in durable, acoustically optimised plywood cabinets with a scratch-resistant polyurea coating. All feature switchable D-XSUB processing for extended low-frequency reproduction down to 32–38Hz. The CXR mk3 and CXS mk3 series are unpowered versions of the DXR/DXS range, primarily intended for installation, fixed system and corporate event applications, such as situations where nearby power sources may be limited. The SWX3220-30TCs and SWX2322P-30MC switches comes with support for 100/25 Gigabit and Multi-Gigabit Ethernet. The SWX3220-30TCs 100G/25G standard L3 switch ensures stable, low-latency transmission of audio and video signals. Fibre connections provide noise resistance and long-distance coverage. An IEEE 1588 PTPv2 Boundary Clock (BC) enables the construction of highly accurate media synchronisation environments. The SWX2322P-30MC 100G/25G intelligent L2 PoE switch can be used to distribute between FOH and stage-side locations, allowing flexible signal routing and high-density encoder/decoder connections with PoE++. In corporate facilities, it supports AV systems spanning multiple floors or departments. The manufacturer has also expanded ProVisionaire Cloud,
Apr 10, 2026 · via proavl-asia.com
Drug dealer busted after sending picture of new 'Turkey teeth' to friend A drug dealer who was caught after sending a photo of his "Turkey teeth" to a friend has been jailed for four years. Coran Davies, 23, sent the picture to his friend, whose phone had been confiscated due to a separate police investigation, from a dentist's chair in Turkey after getting new teeth, alongside messages about supplying cocaine and cannabis. He was identified using facial recognition and arrested in March from his car, alongside passenger Dale Howell, 25, both from Porth, Rhondda Cynon Taf. Officers discovered drugs and thousands of pounds in cash in their possession. Both pleaded guilty to multiple offences and were sentenced on Thursday at Merthyr Crown Court. When Davies sent the picture to his friend, he did not know that their phone was under police analysis. Then on 7 March, officers stopped Davies driving in Tonypandy and arrested him and passenger Howell. At Merthyr Magistrates Court on 9 March, Davies pleaded guilty to multiple offences, including cannabis supply, driving without insurance or a licence, possessing criminal property, and intent to supply cocaine. Howell pleaded guilty to two offences, including possessing criminal property and intent to supply cocaine. Davies was jailed for 48 months and Howell was jailed for 36 months.
Apr 10, 2026 · via bbc.com
Facial recognition has quietly moved from science fiction into everyday life. Whether you’re unlocking your phone, verifying identity in a banking app, or running a reverse face search online, this technology is working behind the scenes—fast, precise, and increasingly intelligent. But here’s the thing: not all facial recognition tools are built the same. Some are designed for enterprise-grade security, others for developers, and a growing number focus on face search and online identity tracking. What is Facial Recognition Technology? Facial recognition is a form of biometric authentication that identifies a person by analyzing their facial features. It works by capturing an image, extracting key facial points (like eyes, nose, and jaw structure), and converting them into a unique digital signature—often called a faceprint. Top 7 Face Search Tools in 2026 (Tested & Compared) Face search technology has evolved rapidly. What started as simple reverse image lookup has now become AI-powered facial recognition capable of identifying individuals across billions of web pages. FaceCheck.ID – Best for Accuracy & Identity Verification FaceCheck.ID consistently ranks as one of the most accurate face search tools available today. Key Features - Detects faces from low-quality, blurred, or masked images - Finds matches across news, social media, and public records - Provides direct source links - Flags suspicious matches (scams, fake identities) Why It Stands Out - Scored 79/80 in real-world testing, outperforming competitors - Excellent for catfish detection and OSINT investigations Best For - Journalists, investigators, and identity verification 2. PimEyes – Best for Large Image Database PimEyes is one of the most well-known facial recognition tools, powered by a massive index of publicly available images. Key Features - Searches billions of web images - Fast results (often under 10 seconds) - Reverse face search with similarity scoring Pros - Strong performance on clear, front-facing
Apr 10, 2026 · via mexc.co
According to multiple reports, University of South Florida assistant coach Oliver Antigua has resigned. ESPN reported earlier today that the NCAA investigating the USF men’s basketball team for possible academic fraud issues. Shortly after that story broke, the news of Antigua’s resignation was confirmed by school officials. South Florida also released the following statement:
“The University of South Florida and the NCAA enforcement staff are working together to investigate and resolve an inquiry into potential violations of NCAA bylaws and university standards by one of our intercollegiate athletic programs.”
“Because the University of South Florida is committed to protecting the integrity of the investigation and ensuring those involved receive fair treatment, we cannot provide any details about the investigation at this time.”
Antigua, the brother of USF head coach Orlando Antigua, had been on the Bulls staff for the past two seasons. He has also worked as an assistant at Seton Hall and Manhattan.
More on this story as it develops.
Apr 10, 2026 · via hoopdirt.com
Killjoys and scatterbrains might have propelled the evolution of the human species. This is, essentially, the theory proposed by linguist Ljiljana Progovac in a new paper published in PNAS Nexus. Progovac argues that clever verb-noun compounds like killjoy—which has a bit more punch than joy killer—were the earliest forms of verbal wit and helped the species survive. They enabled our ancestors to both soothe tempers with humor and compete with words rather than with fists. The wittier the human, the more likely that human would survive. One point of evidence: In brain scans, Progovac and colleagues have found that these compound words produce a stronger effect in a part of the brain called the fusiform gyrus than when the words are separated—particularly a part of the brain thought to be responsible for visual processing and metaphor as well as facial recognition. I spoke with Progovac about so-called spin-buts and burst-cows, about why she calls these compound words living fossils, and how she arrived at her theory. We also talked about how verbal cleverness relates to intelligence and the election of leaders, and whether she thinks of herself as witty. What’s the wittiest thing anyone has ever said to you? I can’t answer that question. These things, they’re in the moment, they happen and they come and they go, and you don’t necessarily even remember them. But there are certain compound words that I find especially witty. There are hundreds, thousands of them, actually. They’re also transient. They were created, used for a while, and then they disappeared. They go out of use. One of the reasons is that many of them are obscene. They refer to body parts and body functions. What were the origins of these compounds? One thing that I noticed, which I find really compelling, is compounds
Apr 10, 2026 · via nautil.us
A new (8 April 2026) rapid response report from the Parliamentary Office of Science and Technology examines the issue of facial recognition technology in policing. UK police forces have used facial recognition technology for around a decade now and the report asks the key questions: - How does it work? - How is it used? and - What are the opportunities and concerns? What is facial recognition? While we might all think we know the answer to this question; it is very helpful to be walked through the process with clarity. Facial recognition technology (FRT) is a biometric technology that estimates the degree of similarity between two faces. UK police forces use FRT to help identify people. FRT identifies people by checking an image of a person against a list of known individuals to find matches. FRT uses artificial intelligence (AI) and machine learning to train systems to recognise faces. FRT speeds up identification and frees up police time. FRT can be broadly organised into three steps: - Input and facial detection: an image of a face is captured and uploaded. - Features extraction: facial features are extracted and translated into a numerical template. - Classification: the numerical template is used to create a ‘similarity score’ with other numerical templates in a database to verify or identify a person. Similarity scores indicate the similarity between someone’s numerical template and a numerical template in a database. Types of numerical templates in databases include ID records or people wanted by the police. The infographic below shows the process. Three types of facial recognition technology used by the police UK police forces primarily use FRT software from private sector companies including NEC, Cognitec and Idemia. Police use three types of FRT: - Retrospective Facial Recognition (RFR): RFR is used across UK police forces
Apr 10, 2026 · via russellwebster.com
Abstract Stampede incidents in densely populated environments remain a critical challenge for public safety, resulting in severe casualties and significant disruption. This paper presents a novel, data-driven framework for automated stampede detection and crowd risk classification, leveraging the integration of Farneback optical-flow computation with a hybrid CNN-LSTM architecture. Two complementary datasets, UCSD Anomaly detection dataset and Agoraset dataset, were combined to capture a broad spectrum of crowd behaviors and densities. The proposed system classifies crowd states into four distinct risk levels: normal, moderate, dense, and risky, thereby offering a more granular assessment than traditional binary models. The model was trained and evaluated on a balanced dataset of 10,000 annotated frames, with rigorous preprocessing and augmentation to ensure robustness. Experimental results demonstrate an accuracy of 99.75%, further cross-dataset evaluation on the UMN benchmark assesses the robustness under domain shift conditions. While the approach shows strong potential for real-time deployment in public event management and emergency response, current limitations include computational latency and challenges in ultra-dense, occluded scenarios. Similar content being viewed by others Data Availability The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request. References Cob-Parro, A.C., Losada-Gutiérrez, C., Marrón-Romera, M., Gardel-Vicente, A., Bravo-Muñoz, I. & Sarker, M.I. A proposal on stampede detection in real environments. In: Proceedings of the IPIN 2021 WiP, Lloret de Mar, Spain (2021). https://ceur-ws.org/Vol-3097/paper29.pdf Sharif, M. H. & Djeraba, C. An entropy approach for abnormal activities detection in video streams. Pattern Recogn. 45(7), 2543–2561. https://doi.org/10.1016/j.patcog.2011.11.023 (2012). Duives, D. C., Oijen, T. & Hoogendoorn, S. P. Enhancing crowd monitoring system functionality through data fusion. Sensors 20(20), 6032. https://doi.org/10.3390/s20216032 (2020). Lalit, R. & Purwar, R. Crowd abnormality detection using optical flow and glcm-based texture features. J. Inf. Techn. Res. 15 (2022) https://doi.org/10.4018/JITR.2022010110 Jadhav, C., Ramteke, R. & Somkunwar, R.
Apr 10, 2026 · via nature.com
Distracted driver and seatbelt detection cameras target drivers who use portable devices while driving. They also can pick up drivers and front seat passengers not wearing their seatbelt. Illegally using a portable device, such as a mobile phone, when driving significantly increases the risk of being involved in a serious crash. In Victoria, widespread roll-out of automated mobile phone enforcement is predicted to prevent 95 casualty crashes per year. What is a portable device? A portable device is any electronic device that can communicate wirelessly and display information but is not an inbuilt device, mounted device, motor bike helmet device or wearable device. Examples of portable devices include: - a dispatch system - a tablet or iPad - a mobile phone - a media player - a camera - a laptop - an information, navigation or entertainment system - a video game console. CB radios and other two-way radios are not included. For full definitions see Road Safety Road Rules 2017. How the cameras work Each camera trailer has 2 cameras and an infra-red flash that are enabled with artificial intelligence (AI) software. The cameras take high-resolution images any time of the day or night, and in all traffic and weather conditions. The AI technology automatically reviews each image. If it does not see a potential offence, it will reject the image. If the AI detects a driver who may be using a portable device or not wearing their seatbelt it flags the image for further review. Images where there might be a potential offence are then checked and verified by qualified independent officers. Illegally using a mobile phone when driving significantly increases the risk of being involved in a serious crash. Images captured by the cameras If you receive an infringement, you can view the images captured at fines.vic.gov.au.
Apr 10, 2026 · via vic.gov.au
Abstract Group activity recognition requires a holistic understanding of individual actions, their spatial relationships, and the surrounding environment. Traditional methods that focus solely on isolated movements often fail to capture the complex inter-player and scene-level dependencies inherent in sports and crowd scenarios. In this research work, a model for group activity recognition is developed. The proposed model combines various contextual features through the integration of poses of individual actors in the scene with the pose-aligned spatial scene context for relational reasoning. Pose features of individual actors are extracted using mmPose, while the scene-level context is encoded through pose-conditioned spatial feature aggregation rather than explicit semantic segmentation. These pose and scene context features extracted are combined and used to construct Actor Relation Graphs (ARGs) using Zero Normalized Cross Correlation (ZNCC) which improves robustness to appearance and variations in illumination. Further, Graph Convolutional Networks (GCNs) are modelled using relationships between individual actors in a scene and their group activities. The proposed framework explicitly combines pose-level and scene-level contextual features into a single relational graph, in contrast to previous ARG-GCN approaches that mainly rely on appearance features. The model is evaluated on two benchmark datasets: the Collective Activity dataset (CAD) and the Volleyball dataset (VD). The model exhibits classification accuracies of 95.02% and 94.81% on CAD and VD, respectively. On a TITAN-XP GPU, the average time per video clip with 41 frames is approximately 0.2 s. The results show that the combination of pose and scene contexts features enhances graph-based relational learning and improves recognition accuracy. Similar content being viewed by others Data availability The Volleyball dataset is publicly available in the “mostafa-saad/deep-activity-rec **”** repository, [https://github.com/mostafa-saad/deep-activity-rec? tab=readme-ov-file#dataset](https:/github.com/mostafa-saad/deep-activity-rec? tab=readme-ov-file) and Collective Activity Dataset is publicly available in the [Computational Vision and Geometry Lab (CVGL) website at Stanford University](https:/cvgl.stanford.edu/projects/collective/collectiveActivity.html) , [https://cvgl.stanford.edu/projects/collective/collectiveActivity.html](https:/cvgl.stanford.edu/projects/collective/collectiveActivity.html) . The datasets used
Apr 10, 2026 · via nature.com
HYDERABAD: The Telangana Higher Education department has planned to make a facial recognition-based attendance system mandatory in all government and private degree colleges from the upcoming academic year to ensure accurate attendance tracking and enhance student welfare. The decision was taken during a project monitoring unit (PMU) meeting chaired by Education Secretary Yogita Rana on Thursday to review key aspects of admissions, attendance and data management in higher education institutions across the state. One of the key decisions taken at the meeting was that the Degree Online Services Telangana (DOST) notification for the 2026-27 academic year will be issued only after the announcement of Intermediate results. In a move to strengthen monitoring, the meeting made it mandatory for both government and private colleges to implement a facial recognition-based attendance system, aimed at ensuring accurate tracking and improving student welfare. The PMU also directed all institutions to strictly enforce mandatory attendance. Universities have been asked to issue circulars to affiliated colleges to ensure compliance, with a focus on improving academic discipline and learning outcomes. During the meeting, TGCHE chairman Balakista Reddy highlighted the introduction of new and innovative undergraduate courses from the 2026-27 academic year, stating that extensive consultations were held with vice-chancellors, subject experts and other stakeholders. He said colleges will be required to enter into MoUs for student internships, while syllabus structures will be provided by universities, along with AI-enabled, tutor-supported learning materials to be made available online. Addressing admission-related concerns, the meeting resolved that students in colleges where DOST admissions are 15% or less will be allowed to opt for sliding to other institutions through the platform to secure better opportunities.
Apr 10, 2026 · via newindianexpress.com
Street photography was built on proximity, on the unscripted moment when two strangers briefly shared the same space and the same gaze. In a world where every face is searchable, traceable, and legally accountable, that proximity no longer carries the same meaning. The future of street photography doesn't look like a sudden collapse. It looks like a slow, managed retreat. On the surface, the genre seems healthy: cameras are easier to use, and cities are more crowded than ever. But the shift isn't about aesthetic trends. It's about a fundamental change in the status of the human face. Street photography was built on the "friction of the second" — the unscripted encounter between strangers. In this tradition, the face wasn't an accessory; it was the raw material of the craft. Now, that material has become toxic. The idea of the "random passerby" is eroding as a category. In a data-driven environment, the person in the frame is no longer anonymous by default. They are a data subject. The language shifts quietly, but the implications are significant: what once appeared as incidental presence now carries traceability. The camera no longer captures a passerby. It captures a profile. From Eye to Scanner The street is no longer a neutral stage. Today, a camera in a public space functions less like an artist's eye and more like a biometric scanner. In many jurisdictions, the moment a face is captured, it ceases to be "character" and becomes "personal data." Once an image is uploaded, licensed, or even used to build a photographer's online brand, it enters a commercial context that strips away the old protections of "artistic freedom." This isn't just a theoretical threat. Reverse-image search has turned every street portfolio into a searchable database. A stranger in your frame is now one click
Apr 9, 2026 · via fstoppers.com
Abstract Image restoration is a vital research area in computer vision, focusing on reconstructing high-quality clear images from degraded observations. Common types of degradation include noise and blur, which may stem from imaging device limitations, environmental interference, and other factors. This paper centers on the design and optimization of multi-stage image restoration networks, conducting in-depth exploration of feature extraction, feature fusion, attention mechanisms, and their practical applications. A multi-stage hybrid attention mechanism-based image restoration network is proposed. Initially, each stage progressively extracts and restores image features. Then, an adaptive feature fusion block enables effective cross-stage information transfer. Finally, by calculating losses at each stage and assigning different weights, the network achieves stable convergence during training. The hybrid attention mechanism enhances the model’s focus on critical features and improves its understanding of the overall image structure. Outstanding performance has been achieved in both image deblurring and denoising tasks. On the GoPro dataset, the restored results achieved a PSNR of 33.26 and an SSIM of 0.963. On the SIDD dataset, the restored results reached a PSNR of 40.23 and an SSIM of 0.963. Furthermore, ablation experiments demonstrated the effectiveness of the multi-stage model, hybrid attention mechanism, and adaptive feature fusion block. Similar content being viewed by others Data availability The datasets generated and/or analysed during the current study are available in the public repositories listed below. For image de-blurring: the GoPro blur dataset (3214 images, 1280 \(\times\) 720 px) was downloaded from https://github.com/SeungjunNah/DeepDeblur_release. The originally provided train/test split (2103/1111 images) was adopted. High-resolution images were cropped into 512 \(\times\) 512 px patches to accelerate training and inference. For image de-noising: the Smartphone Image Denoising Dataset (SIDD) was obtained from https://www.eecs.yorku.ca/ kamel/sidd/. It contains 31888 noisy/clean image pairs; we used the standard split (30 608 training and 1 280 validation images) after per-image
Apr 9, 2026 · via nature.com
A field built long before the boom Artificial intelligence refers to methods that let machines perform tasks linked to human cognition, including pattern recognition, learning, and decision-making. Early systems relied on rule-based programs, which engineers designed to follow fixed instructions in controlled environments. While those systems worked well within narrow limits, they broke down once inputs became unpredictable. As researchers confronted those limits, they shifted toward machine learning in the late 20th century, training algorithms on large datasets rather than encoding rules directly. This transition improved performance in areas like speech recognition and image classification. As computing power expanded, models grew more complex, which in turn allowed deep learning systems to process raw data through layered neural networks. Cycles that shaped today’s moment Although progress appears steady in hindsight, AI research has moved through repeated cycles of optimism and retreat. Funding surged when results looked promising, then declined when systems failed to meet expectations. During those quieter periods, researchers refined techniques and built the groundwork for later advances. In that context, the current wave looks less like a sudden break and more like an acceleration of earlier work. Generative models reached the public after years of incremental gains in data processing and neural network design. Their rapid adoption reflects not only technical progress but also sustained corporate investment. Generative systems under pressure Generative AI systems produce text, images, and code by modeling statistical relationships within massive datasets. Rather than reasoning through problems, they predict outputs that match patterns seen during training. This approach allows them to generate fluent language and detailed content across many domains. At the same time, that strength exposes a consistent weakness. These systems can produce errors that sound convincing, which has raised concerns in fields where accuracy is paramount. Researchers continue to test reliability in areas
Apr 9, 2026 · via futura-sciences.com