In the journey to an autonomous driving future, one of the primary difficulties that remain is the capacity for cars to make rapid, potentially life-saving decisions.\nInnovators are still perfecting this element, and a new research project between CERN and car safety company Zenseact has been published investigating the use of deep learning algorithms to help cars avoid collisions.\nThe technology relies on computer vision to analyze and respond to a car’s environment rapidly and accurately.\n\"Deep learning has strongly reshaped computer vision in the last decade, and the accuracy of image-recognition applications is now at unprecedented levels,” said Christoffer Petersson, Zenseact research lead.\n“Results show great promise for future speed and accuracy increases in image recognition for autonomous vehicles, helping to improve cars’ ability to avoid accidents,” Zenseact said.
"}}}},{"node":{"postTitle":"Adversarial machine learning 101: A new cybersecurity frontier","slug":"adversarial-machine-learning-101-a-new-cybersecurity-frontier","publishDate":"2023-01-31 12:22:25+01:00","createdLocal":"2023-02-01 14:30:57.528496","feedName":"Image Recognition","postSummary":{"childMarkdownRemark":{"html":"Before explaining adversarial machine learning examples, it’s crucial to comprehend how machine learning algorithms process images and videos.\nThe aim of adversarial machine learning projects is to protect machine learning models from malicious manipulation or exploitation by adversarial actors.\nAdversarial machine learning defensesThe two most effective techniques for training AI systems to withstand adversarial machine learning attacks are adversarial training and defensive distillation.\nThe adversarial machine learning example is designed to appear “normal” to humans but causes misclassification by the targeted machine learning model.\nAdversarial machine learning represents a crucial aspect of this approach, leveraging the power of machine learning to defend against evolving cyber threats.
"}}}},{"node":{"postTitle":"Grab appoints Philipp Kandal as Chief Product Officer","slug":"grab-appoints-philipp-kandal-as-chief-product-officer","publishDate":"2023-01-31 00:00:00","createdLocal":"2023-02-01 14:30:57.087267","feedName":"Connected Car","postSummary":{"childMarkdownRemark":{"html":"Grab Holdings Limited, the Southeast Asia’s leading superapp, has on Monday announced that it has promoted Philipp Kandal to Chief Product Officer, effective February 1, 2023.\nGrab said in a statement that Kandal will oversee the product, design and analytics teams and lead the product vision and strategy for Grab.\nAs Chief Product Officer, Philipp will lead teams responsible for delivering products that consumers, driver- and merchant-partners trust and love, while driving profitable growth for the company.\nHe will continue to lead the Geo organization in addition to his Chief Product Officer responsibilities.\nGrabMaps powers Grab services in all eight countries it operates in, and helps Grab avoid tens of millions of dollars in potential third-party mapping fees every year.
"}}}},{"node":{"postTitle":"All You Need to Know about AI and Cryptocurrency in 2023","slug":"all-you-need-to-know-about-ai-and-cryptocurrency-in-2023","publishDate":"2023-01-31 17:00:45+00:00","createdLocal":"2023-02-01 14:30:56.427109","feedName":"Image Recognition","postSummary":{"childMarkdownRemark":{"html":"Experts predict the fusion of AI and Cryptocurrency in 2023 citing the development of AI in a variety of fields.\nWith this development, there are currently rumors regarding AI in cryptocurrency, a much-needed combo.\nExperts anticipate that using AI in current cryptocurrency projects would make greater, effective, and secure solutions.\nConsidering the growth of AI in a range of sectors, experts anticipate the combination of AI and Cryptocurrency in 2023.\nA good overall outlook on a cryptocurrency can predict price gains whenever it relates to investing in cryptocurrencies, whereas a massively negative approach can predict price drops.
"}}}},{"node":{"postTitle":"Mahindra’s XUV400 crosses 10,000 mark for bookings","slug":"mahindras-xuv400-crosses-10000-mark-for-bookings","publishDate":"2023-02-01 04:24:52+00:00","createdLocal":"2023-02-01 14:30:55.585509","feedName":"Connected Car","postSummary":{"childMarkdownRemark":{"html":"February 01, 2023 09:54 am | Updated 09:54 am ISTMahindra claims they have crossed over 10,000 bookings for its XUV400 electric SUV, ever since bookings opened from January 26.\nMahindra will prioritise deliveries for the higher-spec XUV400 EL variant from March this year.\nPrices for the XUV400 currently range between ₹15.99 lakh and ₹18.99 lakh (ex-showroom), and are only applicable for the first 5,000 bookings for each variant of the XUV400.\nThe former is available with the smaller 34.5kWh battery pack with either a 3.3kW or 7.2kW AC charger.\nThe top-spec XUV400 gets features such as a 7.0-inch touchscreen with Mahindra’s AdrenoX software, a single-pane sunroof and connected car technology with over-the-air (OTA) updates.
"}}}},{"node":{"postTitle":"LG to hold global competition on AI image captioning","slug":"lg-to-hold-global-competition-on-ai-image-captioning","publishDate":"2023-02-04 00:00:00","createdLocal":"2023-02-01 14:30:55.532569","feedName":"Image Recognition","postSummary":{"childMarkdownRemark":{"html":"LG AI Research Center at LG Science Park in southwestern Seoul / Courtesy of LGBy Kim Jae-heunLG's artificial intelligence (AI) researchers are moving quickly to lead the convergence trend of visual AI and language AI.\nThe company's research center will hold the LG Global AI Challenge for AI researchers around the world from Feb. 1 until the end of April, LG said Tuesday.\nTo hold the competition, LG will establish a joint research center with Seoul National University's AI Graduate School and U.S. stock image provider Shutterstock.\nLG Global AI Challenge will be held online to evaluate how accurately an AI understands and explains images presented to it, under the theme of \"zero-shot image captioning.\"\nAs zero-shot image captioning advances over time and an AI's image recognition accuracy improves, LG believes the technology will have real-life applications.
"}}}},{"node":{"postTitle":"Tesla gets Justice Department subpoena for self-driving cars","slug":"tesla-gets-justice-department-subpoena-for-self-driving-cars","publishDate":"None","createdLocal":"2023-02-01 14:30:55.106460","feedName":"Connected Car","postSummary":{"childMarkdownRemark":{"html":"Citroen eC3: Company aims to become major EV player in India | TOI AutoThe U.S. Justice Department has requested documents from Tesla Inc .\nTesla is already facing multiple investigations by the National Highway Traffic Safety Administration for problems with its two driver-assist systems, Autopilot and \"Full Self-Driving.\nTeslas using \"Full Self-Driving\" can navigate roads in many cases, but experts say the system can make mistakes, which even CEO Elon Musk acknowledges.\nA separate probe into Teslas that were using Autopilot when they crashed into emergency vehicles started in August 2021.\nSince 2021, Tesla has been beta-testing \"Full Self-Driving\" using owners who haven't been trained on the system but are actively monitored by the company.
"}}}},{"node":{"postTitle":"AerialWaste dataset for landfill discovery in aerial and satellite images","slug":"aerialwaste-dataset-for-landfill-discovery-in-aerial-and-satellite-images","publishDate":"None","createdLocal":"2023-02-01 14:30:53.570389","feedName":"Image Recognition","postSummary":{"childMarkdownRemark":{"html":"The location database is used to extract the positive images from the three data sources.\nA subset of the positive images is enriched with segmentation masks with the help of the ODIN annotator tool.\nWorldView-3 (WV3): high-resolution pan-sharpened RGB images acquired by a commercial satellite sensor (no pan-sharpened near infrared images were used).\nThe evidence attribute is present in most samples, and the severity and area type annotations are specified for ≈72% of the samples.\n5 Distribution of images in the dataset by the (a) evidence, (b) severity and (c) area type properties (n/s = not specified).
"}}}}]}},"pageContext":{"limit":8,"skip":6480,"homeNumPages":1077,"currentPage":811}}}