Intel Labs Pittsburgh Open House, Thursday, November 12, 2009
Download the Intel Labs Pittsburgh 2009 Open House Brochure
Please join us for our Open House on Thursday, November 12, 2009 from 3pm to 6:00PM!
This annual event is an opportunity to experience the full range of our research activities and collaborations, via posters, demonstrations, and direct conversation with the researchers and students involved. We'll be showing our latest work with Carnegie Mellon, University of Pittsburgh, and many others, in personal mobile robots, computer-assisted medicine, brain activity decoding, automatic fast food recognition, programmable matter, parallel machine learning, video-based gesture recognition, neighborhood-aware networking, power efficient server design, cloud computing on big data, cache-savvy algorithms, using solid-state disks for databases, multicore system design, and optical interconnects for datacenters.
Please join us to discuss new research and innovation (ours and yours!), over light refreshments and snacks. We look forward to welcoming you on Nov 12th!
Schedule of Events
Note: Events in blue are by special invitation. All other events are free and open to the public.
Wed, Nov 11, 2009 6:00-9:00pm Welcome reception (invitation only) for researchers and out-of-town visitors.
Beverages served at 6pm. Food served at 7pm.
Location: Schenley Park Cafe, 101 Panther Hollow Road, Pgh, PA 15213, (412) 687-1800 ( Map )
Thu, Nov 12, 2009
12:00-3:00pm Private showing (Invited Intel and university guests only). Lunch served at noon. 3:00-6:00pm Public Open House (open to the public) Note: All events (except the reception on Wednesday evening) are at Intel Labs Pittsburgh on the CMU campus, CIC Bldg, Suite 410, 4720 Forbes Ave, Pgh, PA,15213
Example Demos
Here is a small sampling of the demos you'll be seeing:
- Robot butler - HERB is an autonomous mobile manipulator robot that combines our latest research in object recognition, navigation, motion planning, and grasping to perform complex (for a robot!) tasks around the house. In the demo, you will go to a serving table and use a cell phone to order a drink. The robot will retrieve your drink, bring it to the serving table, and clean away any empties. (with CMU Robotics Institute, Quality of Life Technology Center, and Intel Labs Seattle)
- Brain activity decoding - Researchers have developed techniques that can predict a person's thoughts (read their minds!) from fMRI and MEG images of brain activity with surprising accuracy. In this demo, you'll compete with the computer to see who can do a better job of decoding brain activity. (with CMU SCS and UPitt)
- Food recognition - An interactive Android phone app that allows you take a picture of a fast food menu, and then get nutritional information about each food item in the photo. In this demo, you'll use the Android phone to get nutritional information from printed menus from real fast food restaurants. You'll also be able compete with the computer to see who can do a better job of identifying photos of real fast food items.
- SLIPstream - Play with your computer by pointing at it! (with CMU Robotics Institute and CMU Informedia Project)
- The Nomenclator - See how mobile face recognition technology can be used to give you information about the people you meet. For example, what is their name? Where do you know them from? The device we've all been waiting for! (with Intel Labs Seattle and Intel Labs Berkeley)
- Dynamic Physical Rendering - Hold millimeter-scale prototypes of the building blocks for shape-shifting materials. See new, tiny electrostatic actuators bonded to intricate control circuitry, with the resulting unit similar in size to a grain of rice. Experiment with new approaches to self-reconfiguration of prismatic (square-cornered) shapes. And learn how the programming methodologies we have developed to control shape-shifting ensembles with millions of nodes could also be applied to program and control the highly parallel computing systems of tomorrow. (with CMU SCS and ECE)
- Computer-assisted medicine - Computer-assisted medicine - Computer systems that help doctors diagnose eye diseases, and automatically track thousands of stem cells for days as they move, divide, differentiate, and die. (with CMU Robotics Institute, UPitt Medical School, and Georgia Tech).
- Fast Array of Wimpy Nodes (FAWN) - This demo will show how a small cluster of low-power Atom-based nodes combined with flash memory can serve a real Twitter workload at a fraction of the power consumed by traditional servers. (with CMU SCS)
- Cloud computing - You'll see a live demo of the Intel Open Cirrus cluster. Open Cirrus is a worldwide testbed for cloud computing, with sites at Intel Labs Pittsburgh, HP Labs, Yahoo!, University of Illinois, MIMOS (Malaysia), IDA (Singapore), ETRI (South Korea), Russian Acadamy of Sciences, and Karlsruhe Institute of Technology in Germany. (with HP Labs and Yahoo!)
- Tashi - A cluster management system for cloud computing on big data. (with CMU PDL and the Apache Software Foundation)
- Better video calls from your home - In this demo you'll experience the improvement in video quality you could get from your home if you were able to use your neighbor's access points to relay traffic over multiple broadband connections. (with CMU SCS)
- Neighborhood-aware networking DVR - In this demo you'll be introduced to the concept of a neighborhood-wide Digital Video Recorder that is able to display video content beyond its storage capacity. This becomes possible through the use of high-capacity last mile technologies, like MoCA and WiFi, that allow for the interconnections of multiple DVRs at high speed throughout the neighborhood. (with CMU SCS)
- Using light to create better datacenter networks - Using light to create better datacenter networks. This demo will showcase the performance improvements possible to datacenter applications when using a hybrid architecture that augments the traditional electrical switch tree hierarchy with an optical network. To take full advantage of the high capacity, but circuit-switched optical network, we introduce buffering at the end host that results in increased temporal and spatial traffic skew. We will demonstrate real time OFDM (Orthogonal Frequency Division Multiplexor) digital processing at 8.36 Gb/s using an exhibit prototype of our Optical OFDM transmitter (the fastest to date), and show how adjustable parameters affect the transmission quality of an image. We will also present a reconfigurable optical 4x4 hybrid circuit/packet switch which will form the basis of network testbeds aimed at evaluating photonic interconnect fabrics for next generation data centers (with CMU ECE, CMU SCS, Rice University, University College London, and Columbia University).
- Relational Perception - We illustrate a system which is able to exploit "cheap knowledge", specified as first-order logic formulas, about the relationship between people and objects to vastly improve on person recognition in video.
- Causal Perception - Discovering cause and effect in perception streams requires one to make inferences in highly unstructured noisy environments with very few (maybe even only one) positive examples. We present a system that is a capable of wading through an immense sea of possible causes in an online fashion to identify a small set of plausible candidates for some effect of interest.
Hotel for Open House visitors
Holiday Inn Pittsburgh at University Center
100 Lytton Ave
Pittsburgh, Pa 15213
Front Desk: 412-682-6200
Fax: 412-682-5745The hotel is on the campus of University of Pittsburgh, easy walking distance (3 blocks) from Intel Labs Pittsburgh.
Previous Open Houses