1 / 40

The Next Two Years in Video Surveillance (2010-2011)

The Next Two Years in Video Surveillance (2010-2011). Christian Laforte, President, Feeling Software. Security Canada Central October 21, 2009 Toronto, Ontario. Failed predictions due to wishful thinking. Evolutionary predictions: More of the same. The Bottleneck Approach.

regis
Download Presentation

The Next Two Years in Video Surveillance (2010-2011)

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. The Next Two Years in Video Surveillance (2010-2011) Christian Laforte,President, Feeling Software Security Canada Central October 21, 2009 Toronto, Ontario

  2. Failed predictions due to wishful thinking

  3. Evolutionary predictions: More of the same

  4. The Bottleneck Approach • Near-future developments aim to solve current problems • Key to prediction is to understand these problems and technologies nearly ready for market • There are 5 bottlenecks in Video Surveillance

  5. The 5 Bottlenecks • Video Capture • Video Transfer • Video Storage • Public Acceptance of surveillance • Understanding Video

  6. The 5 Bottlenecks • Video Capture • Video Transfer • Video Storage • Public Acceptance of Surveillance • Understanding Video* *Disclaimer: Feeling Software innovates in this area

  7. Capturing Video Data • From CIF analog to Panoramic Megapixel CCD • Evolutionary: • More megapixels, better dynamic range, cheaper • Disruptive: • 360, panoramic cameras • Night vision PTZ (if enough time, mention fusion as well)

  8. 360 Panoramic cameras • Demo (Immervision Panomorph)

  9. 360 Panoramic cameras • Always records full panorama, unlike Fixed and PTZ • Cost: Fixed < 360 << PTZ • Durability: Fixed > 360 >> PTZ • Effective resolution increasing rapidly, still much lower than high-end PTZ • Most applicable to hallway corners, wide spaces (e.g. Entrance, conference rooms) • Prediction: in 2 years, most large installations will include them to complement fixed and PTZs

  10. Transferring Video • Evolutionary: Faster networks, more widespread Wi-Fi coverage • Disruptive: • Gap between cable and ADSL increasing • Next generation of cellular wireless technologies

  11. Faster broadband • ADSL peaked – limited video surveillance use • Cable still has space to grow: • Download bandwidth will double • Upload bandwidth will quadruple – enough for 16 MPixels/s for less than $100/month • In urban areas, multiple cable connections 5-10 times cheaper than one dedicated T1

  12. Faster broadband for security • Remote monitoring (video, access control) increasingly affordable • Managed services increasingly competitive for SMB • Makes central monitoring for city-wide surveillance more cost-effective

  13. Mobile broadband • Regular 3G is the tip of the iceberg • 3G LTE/4G become widely available in 2010 • Rogers already offers high-speed HSPA • Bandwidth is comparable to Wi-Fi, possibly even faster if you pay extra • Connectivity costs will be high initially, but may decrease rapidly

  14. Mobile broadband for security • Transmit megapixel video from a cell phone, in real-time • Will have a huge, immediate social impact • Will partially displace niche technologies, e.g. mesh networks • Mobile security applications evolving fast • Higher quality video, more features, wider support (e.g. Blackberry) • Demo of Feeling Software’s iGuard

  15. Transferring Video • Evolutionary: Faster networks, more widespread Wi-Fi coverage • Disruptive: • Gap between cable and ADSL increasing • Next generation of cellular wireless technologies • Standardization of IP Cameras

  16. IP camera standards • Current situation: • Myriad of incompatible cameras • ONVIF and PSIA standards only proof-of-concept • 2 years: not enough for complete industry adoption, but ONVIF will likely dominate • Increase cost pressure on camera and VMS • Will have little immediate impact on closed solutions that bundle cameras and software

  17. IP Camera Standards: predictions • In 2011, for cameras we can expect : • ~50% of large camera vendors have good support for standard • All smaller camera vendors will support it • Cheap cameras get 25% cheaper, slightly more powerful • Larger camera vendors differentiate with non-standard features

  18. IP Camera Standards: Impact on VMS software • VMS industry increasingly divided • Very many extremely cheap, limited software • Very few high-end, highly-scalable software • SMBs will gravitate toward the first, very large organizations will need the second • DVRs already in this cycle

  19. Storage • Evolutionary: • H.264 universally supported, no price premium • Latency will go down considerably • Storage cost and size decrease, fewer servers

  20. Public Acceptance of Video Surveillance (Now) • Public fear driven by ignorance and hype • People removing their own privacy barriers: Facebook, Youtube

  21. Public Acceptance of Video Surveillance (soon) • Regular people will send real-time video updates on crimes, disasters in progress • Increase perceived security threat demand higher security • Being watched by everyone  increased surveillance acceptance • Fears of “being watched” decrease, perceived benefits increase • More public cameras, more city-wide surveillance

  22. Understand and Analyze • Capture, transmission, storage: costs decrease and performance increase fast • In contrast, approach to video understanding still very human-intensive, hasn’t changed radically since VCR • Faster, better but doesn’t scale nicely to 100s of cameras

  23. Understand and Analyze • Information overload caused by large installations makes understanding video hard and expensive • Two approaches, complementary • Replace human decision making(Approach most often taken by analytics: very hard, rarely successful) • Optimize human decision making (3D surveillance)

  24. 2011: Analytics finally useful? • License Plate Recognition (LPR)s • Transportation (e.g. traffic flow) • Face Recognition

  25. License Plate Recognition (LPR) • High-end LPRs work very reliably • When installed properly ($) • My wife can attest to this • For law enforcement, very high ROI even with arguably high prices • LPRs here to stay, but expect governments to introduce them gradually to avoid population backlash • Expect cheap, much less effective models in 2011

  26. Overview LPRs Image by D.C.Atty on FlickR

  27. Traffic flow • Generally simpler problem • Movement along very specific paths • Cameras ideally positioned high and looking down • Objects tend to be separate (except car crashes) • Repeating movement patterns (except for weather) • Expect many success stories in the next two years, e.g. DUI arrests

  28. Face Recognition • Every wants the perfect system: knows who everyone is in the picture, all the time, even when they don’t cooperate • Very hard problem, will not be solved by 2011, may never be solved • Best compromise is for specialized, constrained applications • Identification in a constrained lane • Face tracking and identification: Faceflow demo

  29. Overview Some technology still only possible on TV

  30. Understand and Analyze • Information overload caused by large installations makes understanding video hard and expensive • Two approaches, complementary • Replace human decision making(Approach most often taken by analytics: very hard, rarely successful) • Optimize human decision making (3D surveillance)

  31. Live Video + 3D Environment • 3D GIS + 3D projections of video, automatic control of multiple PTZs • Feels like you’re flying a remote-controlled helicopter

  32. Live Video + 3D Environment • See multiple videos in context, navigate across cameras in one mouse-click, always see nearby cameras

  33. Live Video + 3D Environment • Navigate seamlessly across entire cities or floors of a building

  34. 3D video surveillance • Overlay videos in a 3D reproduction of the world • Following a suspect = moving around • See entire floors or city blocks in one screen • Superimpose other information, e.g. position of alarms, personnel, suspects • Compatible with existing equipment and data • Plan and simulate later improvements

  35. 3D video surveillance (2) • Feeling Software’s Omnipresence 3D • More demos online: www.feelingsoftware.com

  36. References • ipvideomarket.info • Ton of information about IP video today and tomorrow • Only $99/year, well worth the money • Press releases: Rogers, Bell, Telus, Videotron • Ask your vendors for roadmaps • If they don’t want to share it with you under NDA, chances are they are not innovating • Keep them accountable • Contact me: claforte@feelingsoftware.com

  37. Q&A and talking points • I will answer question first • Then open room for discussion: • Are other important problems not addressed by these technologies? • Do you agree with my conclusions? • Did I miss strong trends? • Contact me: claforte@feelingsoftware.com

  38. About Feeling Software • Based in Montreal, Canada • Background • Decades of experience in 3D graphics industry • Alumni of ATI (now AMD), EA, Ubisoft, Alias (now Autodesk) • History • 4 years of high-profile 3D consulting • 3D game and film development tools used by 30,000 developers worldwide

  39. Our Clients

  40. NASA-Calibre Software

More Related