Sunday, November 6
| Title |
T01 - Looking at people: The past, the present and the future |
| Speakers |
Leonid Sigal, Thomas Moeslund, Adrian Hilton, Volker Kruger |
| Description |
Over the course of the last 10-20 years the field of computer vision has been preoccupied with the problem of looking at people. Hundreds, if not thousands, of papers have been published on the subject that span face detection, pose estimation, tracking, activity recognition, etc. This tutorial is designed to give an introduction to and assessment of state-of-the-art in this very active field. The tutorial builds on the book: "Visual Analysis of Humans: Looking at People" that will be published by Springer in time for ICCV 2011.
The book is intended to serve the dual purpose of being a reference and a tutorial to the people entering the field. Because thus tutorial is an extension of this idea, it will similarly consists of a series of talks by experts in the corresponding fields. The tutorial will be broken down into 4 parts: (1) detection and tracking, (2) articulated pose estimation and tracking, (3) activity recognition, and (4) applications. In each part we will have 3 invited lecturers. The lectures will be geared towards general CV audience and will outline the key advances and future challenges in the problems involved. |
| URL |
www.cs.brown.edu/~ls/iccv2011tutorial.html |
| Title |
T02 - 3D point cloud processing: PCL (Point Cloud Library) |
| Speakers |
Radu Rusu, Stefan Holzer, Michael Dixon, Vincent Rabaud |
| Description |
With the advent of new, low-cost hardware such as OpenNI compatible cameras and continued efforts in advanced open source 3D point cloud processing, 3D perception gains more and more importance in robotics, as well as other fields. The workshop attempts to motivate new developers and ideas to delve into this subject by offering a tutorial on point cloud processing using the emerging Point Cloud Library (PCL), which presents an advanced and extensive approach to the subject, as well as providing an overview of existing systems applying these techniques. Our goal is to provide an excellent reference material for students and researchers interested in this subject and take our guests through a complete application demonstration (given live) that combines subjects such as filtering, feature estimation, segmentation, registration, object recognition and finally surface reconstruction. The tutorial will be held using OpenNI compatible sensors, so we encourage the audience to bring theirs so we can follow all the steps together. We're assembling a great list of invited speakers that will talk about the usage of PCL in their work and show impressive demos. |
| URL |
www.pointclouds.org/media/iccv2011.html |
| Title |
T03 - Fcam: an architecture and API for computational cameras |
| Speakers |
Kari Pulli, Andrew Adams, Timo Ahonen, Marius Tico |
| Description |
Computational photography has become increasingly popular in recent years, but APIs for controlling cameras have not kept up with the demands of computational photography applications. This issue was addressed by the "Frankencamera'' architecture and its corresponding "FCam'' API, both of which were presented at SIGGRAPH 2010. Since then, researchers, educators, and enthusiasts have begun using FCam to implement their own novel camera applications or teach others to do so. This course will aim to encourage this by going into detail on the rationale behind FCam and how to use it.
The course will demonstrate FCam's use by building from simple examples, working up to a point-and-shoot camera application, and then going beyond that with a sequence of computational photography applications all built using FCam. Thus the course will not only teach FCam, but will also serve as a primer in the basic algorithms in photography and computational photography. The main focus will be on using FCam to develop your own camera applications, but we will also discuss how educators can use FCam in their computational photography courses. |
| URL |
fcam.garage.maemo.org/iccv2011.html |
| Title |
T04 - Variational methods for computer vision |
| Speakers |
Daniel Cremers, Bastian Goldlucke, Thomas Pock |
| Description |
Variational methods are among the most classical and established methods to solve a multitude of problems in computer vision and image processing. Over the last years, they have evolved substantially, giving rise to some of the most powerful methods for optic flow estimation, image segmentation and 3D reconstruction, both in terms of accuracy and in terms of computational speed. Nevertheless, the majority of computer vision researchers is unfamiliar with partial differential equations - possibly because the variational calculus and the subsequent numerical discretization are not part of the classical computer science education. The goal of this tutorial is therefore firstly to familiarize the computer vision community with the basic concepts of variational methods, secondly to point out a number of recent developments including convex relaxation techniques and efficient primal-dual algorithms, and thirdly to provide the audience with hands-on experience of GPU implementation./td> |
| URL |
cvpr.in.tum.de/tutorials/iccv2011 |
| Title |
T05 - Non-rigid registration and reconstruction |
| Speakers |
Alessio Del Bue, Lourdes Agapito, Adrien Bartoli |
| Description |
The tutorial deals with image registration and the 3D reconstruction of deformable shapes using motion as the main visual cue. The aim is to review and discuss a set of general techniques that can be customized given a specific setup. The emphasis will be on the use of physical and statistical shape priors in different imaging scenarios. The tutorial consists of two main parts: image registration and 3D reconstruction from registered images. In the former, we will show how correspondences can be established between images of a deformable shape. In the latter, we will show how these correspondences can be used as inputs to Non-Rigid Structure-from-Motion (NR-SfM) algorithms. The tutorial uses a consistent framework for image registration and 3D reconstruction, which are shown to be two intimately related problems. |
| URL |
www.isr.ist.utl.pt/~adb/tutorial/ |
| Title |
T06 - Learning with inference for discrete graphical models |
| Speakers |
Nikos Komodakis, Pawan Kumar, Nikos Paragios, Ramin Zabih |
| Description |
Several problems in computer vision, pattern recognition, medical imaging and signal processing can be formulated using the discrete graphical models framework. The two main issues faced by researchers when using graphical models are: (i) Learning: How to estimate the parameters of the model?; and (ii) Inference: How to find the best assignment for the variables of the model? In this tutorial we will discuss these two issues, starting from the basics and building up to the state of the art. |
| URL |
www.csd.uoc.gr/~komod/ICCV2011_tutorial/
|
| Title |
T07 - Computer vision fundamentals: robust non-linear least-squares and their applications |
| Speakers |
Pascal Fua, Vincent Lepetit |
| Description |
Least-Squares fitting is one of the fundamental tools of Computer Vision but relatively few young researchers understand their inner workings or are familiar with sophisticated implementations that address the weaknesses of naive ones. This tutorial aims at filling this gap and should therefore benefit young—and maybe not so young--researchers in the field. |
| URL |
cvlab.epfl.ch/~fua/courses/lsq/ |
| Title |
T08 - Geometry constrained parts based detection |
| Speakers |
Simon Lucey, Jason Saragih |
| Description |
The prevalence of non-rigid deformations in real-world objects have presented challenges and opportunities for computer vision researchers for over two decades. Nowhere is this more evident than in HCI applications, where humans are of primary interest, whose facial expressions, hand gestures and the full range of body motion embody intent as well as identifying characteristics useful in recognition. Much work has been aimed at capturing these deformations. Tangential to these efforts is research on the design of recognition algorithms that are invariant to such deformations.
The principal challenges in dealing with object deformations in images is how to account for large variations in appearance an object can undergo due to illumination, viewpoint, and intra-class variabilities, in addition to the non-rigid deformations themselves. In the past few years, separate streams of research on the various types of deformable objects have converged, where unified formal representations and optimisation strategies have led to a paradigm that is, in a way, object invariant. In particular, the most successful setting for this problem that has emerged amongst the various streams is that of geometry constrained parts based detection.
In this tutorial, we will review the history of research on deformable object detection and examine the development of models and algorithms as the state-of-the-art converges to a unified theory. This will include detailed analysis of the famous Active Shape Models and Active Appearance Models and how they relate to the contemporary Pictorial Structure Models. The tutorial will also identify the key challenges and open problems that remain unresolved. |
| URL |
ci2cv.net/tutorials/iccv-2011 |
| Title |
T09 - Decision forests for classification, regression, clustering and density estimation |
| Speakers |
Antonio Criminisi |
| Description |
In this tutorial we will present a general model of decision forests and discuss how it can be used for a large variety of supervised and unsupervised tasks in machine learning and computer vision. Numerous toy examples will help explain and demonstrate how small variants of the basic forest model correspond to powerful algorithms for efficient: classification, regression, density estimation, manifold learning and semi-supervised learning. Details of exemplar real-world applications including human tracking in Microsoft XBox Kinect will be presented at the end. |
| URL |
research.microsoft.com/en-us/people/antcrim/decisionforeststutorial.aspx |
| Title |
T10 - Color image understanding: from acquisition to high-level image understanding |
| Speakers |
Theo Gevers, Keigo Hirakawa, Joost van de Weijer |
| Description |
In this tutorial we address theory and techniques to optimally exploit available color information from the digital camera processing pipeline up to high-level color image understanding. The aim is to provide attendees with basics on color theory and a practical set of techniques which allow to effectively use color information in computer vision applications. We will show that color information is a powerful tool for image understanding. The course is divided into three parts. First, this tutorial addresses key technical challenges to imaging pipeline design presented by demands such as shrinking device footprints, increasing throughput, and enhancing color fidelity. Next, the fundamentals of color image processing will be outlined, such as color representation and reflection models, photometric invariance, color constancy, and color saliency. The third part of the course will focus on the practical usage of color in computer vision applications. We will show examples from a number of relevant application areas, such as edge detection, image segmentation, motion detection and object tracking, object recognition, image and video retrieval, and scene classification. |
| URL |
www.cat.uab.cat/~joost/tutorial_iccv.html |
|
