Call for Participation

 
 

2nd ACM/IEEE International Conference on
Distributed Smart Cameras (ICDSC-08)

September 7-11, 2008
Stanford University, California, USA

Tutorials

Tutorials will be on Sunday September 7, 2008. For tutorials schedule click here.

Tutorials will be held in the Y2E2 building in iRoom. For location maps click here for local map and here for campus map.

  • Gary Bradski
    Learning OpenCV - Computer Vision with the Open Source Computer Vision Library
    • Pre-install OpenCV on your laptop for this tutorial.
    • Instructions for installing OpenCV provided by the tutorial instructor can be found here.
  • James Fung
    Accelerating Computer Vision on the GPU
    • Presentation slides and sample exercises provided by the tutorial instructor can be found here.

Gary Bradski, Stanford, Willow Garage (and formerly Intel)

Learning OpenCV
Computer Vision with the Open Source Computer Vision Library

Abstract
Bring your laptop, hopefully with OpenCV already installed (free from source forge http://sourceforge.net/projects/opencvlibrary/) and we'll go trhough some programming and computer vision basics, up through tracking and on to calibration, ground plane extraction, stereo and on to object recognition. This library is free for commercial use, is back under active development and so should fit your computer vision needs well. A companion text will be available in mid September at http://www.amazon.com/
Learning-OpenCV-Computer-Vision-Library/dp/0596516134
.

Speaker's Biography
Dr. Gary Rost Bradski is a consulting professor in the CS department at Stanford University AI Lab where he mentors robotics, machine learning and computer vision research. He is also Senior Scientist at Willow Garage, a recently founded robotics research institute/incubator. He has a BS degree in EECS from U.C. Berkeley and a PhD from Boston University. He has 21 years of industrial experience applying machine learning and computer vision spanning option trading operations at First Union National Bank, to computer vision at Intel Research to machine learning in Intel Manufacturing and several startup companies in between. Gary started the Open Source Computer Vision Library (OpenCV), the statistical Machine Learning Library (MLL comes with OpenCV), and the Probabilistic Network Library (PNL). OpenCV is used around the world in research, government and commercially. The vision libraries helped develop a notable part of the commercial Intel performance primitives library (IPP). Gary also organized the vision team for Stanley, the Stanford robot that won the DARPA Grand Challenge autonomous race across the desert for a $2M team prize and helped found the Stanford AI Robotics project at Stanford working with Professor Andrew Ng. Gary has over 50 publications and 13 issued patents with 18 pending. He lives in Palo Alto with his wife and 3 daughters and bikes road or mountains as much as he can.

James Fung, NVidia

Accelerating Computer Vision on the GPU

Abstract
In a distributed smart camera network, greater processing power at each node is desirable to achieve video framerate processing of sophisticated algorithms at high resolutions. Recently, researchers in many fields have begun looking to Graphics Processing Units (GPUs) for hardware acceleration. Whereas in the past, algorithms had to be mapped into graphics programming semantics, modern GPUs now provide more flexibility both in terms of architecture and programmability. For smart camera networks, this means that more and smarter processing may be conducted at the camera nodes to produce higher level information, possibly reducing communication requirements while also allowing for faster, high level processing of fused data.

This tutorial will discuss how to program imaging and vision algorithms on the GPU with NVIDIA CUDA. CUDA is an extension to the familiar C programming language that allows C-like functions to be run on the GPU. Topics covered will include current GPU architecture and data-parallel computing using CUDA. Programming methods and architectural features advantageous to image processing and computer vision will be presented.

Speaker's Biography
Dr. Fung's work has been in the area of applying GPU Hardware for parallel general purpose computing, including implementing Computer Vision on the GPU. He is lead author of the OpenVIDIA project, which won the ACM Multimedia 2005 Open Source Software Award. "Computer Vision on the GPU" in the popular GPU Gems 2 series of graphics programming books. This work has achieved implementation of vision algorithms on the GPU, including projective image stitching, Chirplet detection, Radon Transforms and natural feature processing and matching. He has authored over a dozen peer reviewed papers in IEEE and ACM conferences in the areas of parallel GPU Computer Vision and Mediated Reality. He currently works at NVIDIA examining computer vision and image processing on graphics hardware.