ICSI held its annual open house on Thursday, February 14 in conjunction with the Berkeley EECS Annual Research Symposium. Group directors gave overviews of their groups' work, and researchers demonstrated their latest results in network measurement and security, multimedia analysis, computer vision, artificial intelligence, and speech and audio analysis. Below are photos from the event:
Every year, we host an open house in conjunction with the Berkeley EECS Annual Research Symposium (BEARS). This year, BEARS is on Thursday, February 14. Group directors will be giving overviews of their groups' work, and senior researchers will be on hand throughout to talk about and demonstrate their research. The open house starts at 2 p.m. - we hope you can make it!
The abstracts for all our demos are on the events page, but here are a few previews:
At the Open House, we announced two new research groups, one to focus on multimedia analysis and retrieval and one to nurture nascent research efforts that don't fit in elsewhere at ICSI. Read more in the press release.
Four visitors from Finland recently joined our Networking and Security Group: Andrei Gurtov, Dmitriy Kuptsov, Andrey Lukyanenko, and André Schumacher. They are here as part of ICSI's Finnish visiting program, which is funded by the Finnish Funding Agency for Technology and Innovation through Aalto University and the Helsinki Institute for Information Technology.
ICSI is pleased to announce that, as of January 1, 2013, Professor Vern Paxson is the director of Networking and Security research. Paxson was previously a senior researcher in Networking. He takes the role of director over from Professor Scott Shenker, who directs the recently established Research Initiatives area and serves as chief scientist. Networking and Security research scientists will conduct research on Internet security issues, including investigating the underground economy.
ICSI is pleased to announce that, as of January 1, 2013, Steven Wegmann is the new leader of the ICSI Speech Group. Professor Nelson Morgan, the founder and leader of the group for more than 24 years, will be focusing on research activities, passing on the leadership of the team to Steven. After earlier collaboration with Morgan and his team, Steven formally joined the group in 2011. For nearly two decades prior to that, he pursued speech research at companies such as Dragon, VoiceSignal, and Nuance. He received a PhD in mathematics from the University of Warwick, and his early work was in algebraic topology. Steven’s recent research interest is in the analysis of the failings of statistical models commonly used in speech recognition. For more, see the news brief.
You get a friend request. The name is something you remember from high school. The picture looks familiar. The account has pictures of the person on vacation and their relationship status is "in a relationship" with someone else whose name is also familiar. Obviously you know this person, right? Not necessarily - fake accounts on social networks are common, and while it may seem odd to fake being a real person, if enough information about a person is already online, it is easy to do. If you've followed this series of posts, you are probably already aware that a lot of information about people already exists online, even if that person didn't post any of it. So while you might know that your close friends on social networks really are the people they claim to be, if it's not someone you talk to frequently in real life, it's entirely possible the profile on Facebook (or other social networks) is someone who is only pretending to be that person. So you could be sharing information with people you don't know, even if all of your contacts are names that are familiar to you.
There is a lot of information available online - but the reality is that anyone who can type can post something online. It may or may not be true, and it can be very difficult to determine whether or not something you see online is from a reliable source. This applies to photos and videos as well as text - many people think that if they see a picture of something, then it must be real - but photo (and video) editing software can make things look real that in fact are not. Think about a movie with really good special effects, for example. Because of this, it's important to remember that when you post something, true or not, there will be people who believe it and others who don't. In addition, if you post something that is meant to be sarcasm or a joke of some sort, be aware that many people might not see it that way. Be very cautious with the words you choose to use in your posts online, because you never know who might be reading them and taking them at face value.
Using techniques from the field of robotics, Vision Group researchers and their colleagues have developed a method for detecting objects in images that intelligently selects which object detectors to use and which to ignore in order to complete a task within given time constraints. The paper was presented in December at the Neural Information Processing Systems Conference. It’s by Vision Group researcher Sergey Karayev and group leader Trevor Darrell as well as Tobias Baumgartner of RWTH Aachen University and Mario Fritz of MPI for Informatics, who has worked at ICSI as a postdoc.