BEGIN:VCALENDAR
VERSION:2.0
PRODID:IEEE vTools.Events//EN
CALSCALE:GREGORIAN
BEGIN:VTIMEZONE
TZID:Asia/Taipei
BEGIN:STANDARD
DTSTART:19790930T230000
TZOFFSETFROM:+0900
TZOFFSETTO:+0800
TZNAME:CST
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20230331T175045Z
UID:396FD314-FA32-4343-B431-DAC3184A644A
DTSTART;TZID=Asia/Taipei:20230331T193000
DTEND;TZID=Asia/Taipei:20230331T210000
DESCRIPTION:We are at the very beginning of the era of Internet of Video Th
 ings (IoVT)\, where many cameras collect a huge amount of visual data to b
 e analyzed. As the number of cameras and applications grows exponentially\
 , it is critical to use artificial intelligence (AI) to process this data 
 because humans cannot handle it all. However\, designing efficient IoVT sy
 stems poses many challenges\, such as accuracy\, energy efficiency\, and p
 rocessing speed.\n\nTo address these challenges\, this talk will focus on 
 six topics related to the co-optimization of video compression and compute
 r vision algorithms for efficient IoVT systems:\n1. AI-aware compression\,
  which optimizes video compression for machines to consume the video for b
 etter video analytics results\, rather than optimizing for human perceptua
 l comfort.\n2. AI-assisted compression\, which uses AI algorithms to assis
 t the decisions made in compression tools for different types of video\, s
 uch as video-on-demand vs. live broadcast.\n3. AI-based compression\, wher
 e the image/video is compressed using AI algorithms\, such as deep learnin
 g\, instead of commonly used standards.\n4. Compression-aware AI\, which e
 nsures that computer vision algorithms are aware of the potential artifact
 s caused by lossy compression\, improving the accuracy of the video analyt
 ics system.\n5. Compression-assisted AI\, which uses information from the 
 compressed bit streams\, such as motion vectors\, to assist the computer v
 ision algorithms.\n6. Compression-based AI\, which applies computer vision
  directly to compressed domain data\, reducing the decompression time.\n\n
 Overall\, the co-optimization of video compression and computer vision alg
 orithms is critical to designing efficient IoVT systems. This talk aims to
  shed light on the challenges and opportunities involved in this co-optimi
 zation process.\n\nCo-sponsored by: Bor-Sung Liang\n\nSpeaker(s): Yen-Kuan
 g Chen\n\nRoom: B1 (International Conference Hall)\, Bldg: Engineering Bui
 lding 4\, B1 (International Conference Hall)\, Engineering Building 4\, No
 . 1001\, University Road\, National Yang Ming Chiao Tung University\, Hsin
 chu\, T'ai-wan\, Taiwan\, 300\, Virtual: https://events.vtools.ieee.org/m/
 354620
LOCATION:Room: B1 (International Conference Hall)\, Bldg: Engineering Build
 ing 4\, B1 (International Conference Hall)\, Engineering Building 4\, No. 
 1001\, University Road\, National Yang Ming Chiao Tung University\, Hsinch
 u\, T'ai-wan\, Taiwan\, 300\, Virtual: https://events.vtools.ieee.org/m/35
 4620
ORGANIZER:bs.liang@mediatek.com
SEQUENCE:20
SUMMARY:AI and Video Compression in the Era of Internet of Video Things
URL;VALUE=URI:https://events.vtools.ieee.org/m/354620
X-ALT-DESC:Description: <br /><p>We are at the very beginning of the era of
  Internet of Video Things (IoVT)\, where many cameras collect a huge amoun
 t of visual data to be analyzed.&nbsp\; As the number of cameras and appli
 cations grows exponentially\, it is critical to use artificial intelligenc
 e&nbsp\;(AI) to process this data because humans cannot handle it all. How
 ever\, designing efficient IoVT systems poses many challenges\, such as ac
 curacy\, energy efficiency\, and processing speed.<br /><br />To address t
 hese challenges\, this talk will focus on six topics related to the co-opt
 imization of video compression and computer vision algorithms for efficien
 t IoVT systems:<br />1. AI-aware compression\, which optimizes video compr
 ession for machines to consume the video for better video analytics result
 s\, rather than optimizing for human perceptual comfort.<br />2. AI-assist
 ed compression\, which uses AI algorithms to assist the decisions made in 
 compression tools for different types of video\, such as video-on-demand v
 s. live broadcast.<br />3. AI-based compression\, where the image/video is
  compressed using AI algorithms\, such as deep learning\, instead of commo
 nly used standards.<br />4. Compression-aware AI\, which ensures that comp
 uter vision algorithms are aware of the potential artifacts caused by loss
 y compression\, improving the accuracy of the video analytics system.<br /
 >5. Compression-assisted AI\, which uses information from the compressed b
 it streams\, such as motion vectors\, to assist the computer vision algori
 thms.<br />6. Compression-based AI\, which applies computer vision directl
 y to compressed domain data\, reducing the decompression time.<br /><br />
 Overall\, the co-optimization of video compression and computer vision alg
 orithms is critical to designing efficient IoVT systems. This talk aims to
  shed light on the challenges and opportunities involved in this co-optimi
 zation process.</p>
END:VEVENT
END:VCALENDAR