home bbs files messages ]

Just a sample of the Echomail archive

Cooperative anarchy at its finest, still active today. Darkrealms is the Zone 1 Hub.

   EARTH      Uhh, that 3rd rock from the sun?      8,931 messages   

[   << oldest   |   < older   |   list   |   newer >   |   newest >>   ]

   Message 8,578 of 8,931   
   ScienceDaily to All   
   AI that uses sketches to detect objects    
   21 Jun 23 22:30:28   
   
   MSGID: 1:317/3 6493ce6f   
   PID: hpt/lnx 1.9.0-cur 2019-01-08   
   TID: hpt/lnx 1.9.0-cur 2019-01-08   
    AI that uses sketches to detect objects within an image could boost   
   tumor detection, and search for rare bird species    
      
     Date:   
         June 21, 2023   
     Source:   
         University of Surrey   
     Summary:   
         Teaching machine learning tools to detect specific objects in a   
         specific image and discount others is a 'game-changer' that could   
         lead to advancements in cancer detection, according to researchers.   
      
      
         Facebook Twitter Pinterest LinkedIN Email   
      
   ==========================================================================   
   FULL STORY   
   ==========================================================================   
   Teaching machine learning tools to detect specific objects in a   
   specific image and discount others is a "game-changer" that could lead   
   to advancements in cancer detection, according to leading researchers   
   from the University of Surrey.   
      
   Surrey is set to present its unique sketch-based object detection tool   
   at this year's Computer Vision, Pattern, and Recognition Conference   
   (CVPR). The tool allows the user to sketch an object, which the AI will   
   use as a basis to search within an image to find something that matches   
   the sketch -- while discounting more general options.   
      
   Professor Yi-Zhe Song, leads this research at the University of Surrey's   
   Institute for People-Centred AI. He commented: "An artist's sketch is   
   full of individual cues that words cannot convey concisely, reiterating   
   the phrase 'a picture paints a thousand words'. For newer AI systems,   
   simple descriptive words help to generate images, but none can express   
   the individualism of the user or the exact match the user is looking for.   
      
   "This is where our sketch-based tool comes into play. AI is instructed by   
   the artist via sketches to find an exact object and discount others. Which   
   can be amazingly helpful in medicine, by finding more aggressive tumours,   
   or helping to protect wildlife conservation by detecting rare animals."   
   An example that researchers use in their paper to the conference is of   
   the tool helping to search a picture full of zebras -- with only a sketch   
   of a single zebra eating to direct its search. The AI tool takes visual   
   cues into account, such as pose and structure, but bases the decisions   
   off the exact requirements given by the amateur artist.   
      
   Professor Song continued: "The ability for AI to detect objects based on   
   individual amateur sketches introduces a significant leap in harnessing   
   human creativity in Computer Vision. It allows humans to interact with   
   AI from a whole different perspective, no longer letting AI dictate   
   the decisions, but asking it to behave exactly as instructed, keeping   
   necessary human intervention."  This research will be presented at the   
   Computer Vision, Pattern, and Recognition Conference (CVPR) 2023 which   
   showcases world-leading AI research on a global stage. The University of   
   Surrey sees an exceptional number of papers accepted to the CVPR 2023,   
   far above other educational institutions, with over 18 papers accepted   
   and one nominated for the Best Paper Award.   
      
   The University of Surrey is a research-intensive university, producing   
   world- leading research and delivering innovation in teaching to transform   
   lives and change the world for the better. The University of Surrey's   
   Institute for People-Centred AI combines over 30 years of technical   
   excellence in the field of machine learning with multi-disciplinary   
   research to answer the technical, ethical and governance questions   
   that will enable the future of AI to be truly people-centred. A focus   
   on research that makes a difference to the world has contributed to   
   Surrey being ranked 55th in the world in the Times Higher Education   
   (THE) University Impact Rankings 2022, which assesses more than 1,400   
   universities' performance against the United Nations' Sustainable   
   Development Goals (SDGs).   
      
       * RELATED_TOPICS   
             o Matter_&_Energy   
                   # Technology # Detectors # Telecommunications #   
                   Forensic_Research   
             o Computers_&_Math   
                   # Artificial_Intelligence # Communications #   
                   Computers_and_Internet # Neural_Interfaces   
       * RELATED_TERMS   
             o Computer_and_video_games o Full_motion_video o   
             Electron_microscope o Massively_multiplayer_online_game o   
             Game_theory o Data_mining o Neutrino o Radar   
      
   ==========================================================================   
   Story Source: Materials provided by University_of_Surrey. Note: Content   
   may be edited for style and length.   
      
      
   ==========================================================================   
      
      
   Link to news story:   
   https://www.sciencedaily.com/releases/2023/06/230621105422.htm   
      
   --- up 1 year, 16 weeks, 2 days, 10 hours, 50 minutes   
    * Origin: -=> Castle Rock BBS <=- Now Husky HPT Powered! (1:317/3)   
   SEEN-BY: 15/0 106/201 114/705 123/120 153/7715 218/700 226/30 227/114   
   SEEN-BY: 229/110 112 113 307 317 400 426 428 470 664 700 291/111 292/854   
   SEEN-BY: 298/25 305/3 317/3 320/219 396/45 5075/35   
   PATH: 317/3 229/426   
      

[   << oldest   |   < older   |   list   |   newer >   |   newest >>   ]


(c) 1994,  bbs@darkrealms.ca