The Distributed Camera: Modeling the World from Online Photos with Noah Snavely

  • July 25, 2012 · 5:30 PM
  • Google NYC (8th Ave Entrance)

We live in a world of ubiquitous imagery, in which the number of images at our fingertips is growing at a seemingly exponential rate.  These images come from a wide variety of sources, including Google Maps and related sites, webcams, and millions of photographers around the world uploading billions and billions of images to photo-sharing websites.  Taken together, these sources of imagery can be thought of as constituting a distributed camera capturing the entire world at unprecedented scale, and continually documenting its cities, mountains, buildings, people, and events.  This talk will focus on how we might use this distributed camera as a fundamental new tool for science, engineering, and environmental monitoring, and how a key problem is *calibration* -- determining the geometry of each photo, and relating it to all other photos, in an efficient, automatic way.  I will describe our ongoing work on using automated 3D reconstruction algorithms for recovering such geometry from massive photo collections.

Noah Snavely is an assistant professor of Computer Science at Cornell University, where he has been on the faculty since 2009. He received a B.S. in Computer Science and Mathematics from the University of Arizona in 2003, and a Ph.D. in Computer Science and Engineering from the University of Washington in 2008. Noah works in computer graphics and computer vision, with a particular interest in using vast amounts of imagery from the Internet to reconstruct and visualize our world in 3D, and in creating new tools for enabling people to capture and share their environments. His thesis work was the basis for Microsoft's Photosynth, a tool for building 3D visualizations from photo collections that has been used by many thousands of people. Noah is the recipient of a Microsoft New Faculty Fellowship and an NSF CAREER Award, and has been recognized by Technology Review’s TR35.

 

 

 

 

 

 

 

----------

The agenda for this event is:

5:30 - 6:30 pm: Attendees Register / Networking
6:30 - 6:35 pm: Welcome & Intro
6:35 - 7:35 pm: Presentations
7:35 - 8:00 pm: Q/A

Please note the change in venue to the 8th Avenue side of the building.  If you mistakenly go to the 9th avenue entrance, building security will ask you to walk around the building (on the outside!) to the 8th Avenue side.

The RSVP name will be provided to the building security in advance - so please do register with your real name (this will significantly speed up registration). To allow all participants (in the community) to get a clear visibility into their schedules; for all upcoming talks we will make seating reservable one week prior to the event on July 18th at 2:30 pm.

Google volunteers will also be present at the event to answer any questions you may have, look for people who are wearing "Google Wear".

See you there!

 

Join or login to comment.

  • Raymond de L.

    This meetup was extremely interesting and the demonstrated technology quite impressive.

    August 19, 2012

  • Brad H.

    Fantastic presentation and speaker. The only (huge) problem was the constant disruption by audience questions, especially from attendees in the front few rows. Without microphones, their questions were inaudible to attendees seated in the back, and those Snavely repeated through his own mic were frequently clearly addressed later in his presentation (which in turn, seemed to have been cut very short). It would have been much more enjoyable for the vast majority to have had a microphoned Q&A session at the end.

    August 1, 2012

  • Amit P.

    liked it, keep the questions until the end of the talk.

    July 28, 2012

  • Flavio

    Very interesting technology with many dual use applications. Noah was a great speaker and very enthusiastic about the research.

    July 27, 2012

  • Robert G.

    Excellent presentation. Noah did a fine job.

    July 27, 2012

  • Robert G.

    @Peter (and others interested in understanding the full history). The use of photogrammetric methods to discern geometric details from multiple reconnaissance photographs long pre-date the advent of Hitler's V-1 and V-2. A casual check of Wikipedia (at http://en.wikipedia.org/wiki/Ph...­) confirms. The wikipedia page, admittedly without attribution, credits photogrammetry as arising with the advent of photography in the mid-19th century. References litter the historical record.

    July 27, 2012

  • David S.

    @Peter,

    The paper describes the use of extractions of geometry from motion using techniques that were invented in the late 90's and 2000's. This is very different from stereographic alignment between slightly mismatched images of a given scene, it applies in particular graph theoretic concepts for extracting geometry from motion that have seen work in last 15 years (particularly when the foundations of network theory started being explored mathematically). Did you read the paper?

    July 27, 2012

  • David S.

    Thanks Ben, that's the one, downloaded to Drive and is now on my phone for reading during tomorrow's commute.

    @Peter, careful using the word "similar" here. Mathematically there are significant differences between the "synth" generation process and simulated stereography. For one, some of the mathematical techniques used by Noah and his team *didn't exist* in world war II. Your latter point regarding computability is well taken. FYI.

    July 26, 2012

  • David S.

    This particular meetup exemplifies why the Google Tech talks are one of the best in the city. I hadn't researched who Noah was prior to the talk but about 15 minutes in I realized he was describing the "Photosynth" technology that was in the tech. blogs a few years ago. I spent a few hours at Microsoft's site flying through "synths" (the only low point was it was run in Silverlight...oy vey!). Noah was engaging, knowlegable and too willing to entertain interrupting questions from the audience...but one of the best talks I've seen this year. Well done!

    July 26, 2012

  • June K.

    Fascinating work being done by Noah Snavely. I hope Google will invite him to speak again in the future so that we can see the progress of those scale models.

    July 26, 2012

  • Peter L.

    I mentioned to Ben Snavely after the talk, that there was a similar effort during World War 2, that changed the course of the war by saving England from rocket bombardment, using aerial photos in stereo to measure strange new facilities (rocket ramps for a sled, pointing to England). It was done by hand, and is in a great documentary on Nova. http://www.pbs.org/wgbh/nova/mi...­ As Turing noted: anything done on a large computer can done on a small computer/multiprocessed.

    2 · July 26, 2012

  • Matthew D.

    Was informative and enjoyable.

    July 26, 2012

  • Michael M.

    Q&A might have better been left for the end of presentation.

    July 26, 2012

  • Yen

    Exciting opportunities when the real world is fully reconstructed online from user photos. Interesting challenges to accurately automate the calibration process of all those photos and empowering photographers to fill in the gaps.

    July 26, 2012

  • A former member
    A former member

    If you're curious about the algorithm specifics I found this paper (thanks to Google of course) https://www.google.com/url?url=h...­

    2 · July 26, 2012

  • A former member
    A former member

    Really interesting talk. Speaker was engaging, had great materials and brought the topic really to life. I would have liked to see questions held til the end or at least until the end of each section, talk would have flowed better.

    July 26, 2012

  • Jonathan N. W.

    Great topic. Good speaker. Go Big Red!

    July 26, 2012

  • Richard E B.

    It was very interesting !

    July 26, 2012

  • Gabe G.

    REALLY EXCELLENT

    ignoring the people in front would have made it better

    July 26, 2012

  • David S.

    totally agrees with Ben (Chirlin)....but some people up front always consistently interrupt the talks. *sigh*

    1 · July 26, 2012

  • A former member
    A former member

    Good talk. Speaker shouldn't have allowed for questions mid-talk though. Completely broke the flow.

    1 · July 26, 2012

  • Galfromdownunder

    I'm wondering what happens if this gets into the wrong hands. Visions of the Death Star cross hairing Alderaan just crossed my troubled mind ...

    July 25, 2012

  • David D.

    I like the panoramic photo someone uploaded with photosynth. Stay classy ;-)

    July 25, 2012

  • Richard E B.

    Yes tell him to stop laughing at stupid stuff that is not funny.

    July 25, 2012

  • David D.

    Is it just me or is the old guy in the front really annoying?!

    3 · July 25, 2012

  • A former member
    A former member

    Taking photos of this meetup is so meta.

    July 25, 2012

  • Victor A.

    Having gone to previous meet ups, I know that they do have the cameras rolling. Not certain if the footage gets uploaded though.

    July 24, 2012

  • Nik

    Why not consider streaming this event over the web? (similar to the one by NYC Tech meetup)

    July 24, 2012

  • Vijaya K.

    Looking forward to attending this session!

    July 16, 2012

People in this
Meetup are also in:

Imagine having a community behind you

Get started Learn more
Henry

I decided to start Reno Motorcycle Riders Group because I wanted to be part of a group of people who enjoyed my passion... I was excited and nervous. Our group has grown by leaps and bounds. I never thought it would be this big.

Henry, started Reno Motorcycle Riders

Sign up

Meetup members, Log in

By clicking "Sign up" or "Sign up using Facebook", you confirm that you accept our Terms of Service & Privacy Policy