Lab Meeting July 6th (Casey): Live Dense Reconstruction with a Single Moving Camera (CVPR 2010)

Authors: Richard A. Newcombe and Andrew J. Davison

Abstract:

We present a method which enables rapid and dense reconstruction of scenes browsed by a single live camera. We take point-based real-time structure from motion (SFM) as our starting point, generating accurate 3D camera pose estimates and a sparse point cloud. Our main novel contribution is to use an approximate but smooth base mesh generated from the SFM to predict the view at a bundle of poses around automatically selected reference frames spanning the scene, and then warp the base mesh into highly accurate depth maps based on view-predictive optical flow and a constrained scene flow update. The quality of the resulting depth maps means that a convincing global scene model can be obtained simply by placing them side by side and removing overlapping regions. We show that a cluttered indoor environment can be reconstructed from a live hand-held camera in a few seconds, with all processing performed by current desktop hardware. Real-time monocular dense reconstruction opens up many application areas, and we demonstrate both real-time novel view synthesis and advanced augmented reality where augmentations interact physically with the 3D scene and are correctly clipped by occlusions.

[Paper]

Lab Meeting July 6th 2010 (Andi): Upsampling Range Data in Dynamic Environments (CVPR 2010 )

Authors

Jennifer Dolson, Jongmin Baek, Christian Plagemann and Sebastian Thrun (Stanford University)

Abstract

We present a flexible method for fusing information from optical and range sensors based on an accelerated high-dimensional filtering approach. Our system takes as input a sequence of monocular camera images as well as a stream of sparse range measurements as obtained from a laser or other sensor system. In contrast with existing approaches, we do not assume that the depth and color data streams have the same data rates or that the observed scene is fully static. Our method produces a dense, high-resolution depth map of the scene, automatically generating confidence values for every interpolated depth point. We describe how to integrate priors on object shape, motion and appearance and how to achieve an efficient implementation using parallel processing hardware such as GPUs.

paper

Lab Meeting June 29th, 2010 (KuoHuel): People Tracking with Human Motion Predictions from Social Forces (ICRA'10)

Title: People Tracking with Human Motion Predictions from Social Forces

Authors: Matthias Luber, Johannes A. Stork, Gian Diego Tipaldi, and Kai O. Arras

Abstract:

For many tasks in populated environments, robots need to keep track of present and future motion states of people. Most approaches to people tracking make weak assumptions on human motion such as constant velocity and direction. But even over a short period, human motion behavior is more complex and influenced by factors such as an intended goal, other people, objects in the environment, or social rules. Therefore, more sophisticated motion models are highly desirable especially since people frequently undergo lengthy occlusion events.

For the study of crowd behavior or evacuation dynamics, computational models that describe individual and collective pedestrian dynamics have been developed in e.g. the social psychology community. In this paper, we make use of such a model for the purpose of people tracking. Concretely, we integrate a pedestrian dynamics model based on social forces into a multi-hypothesis target tracker. We show how the re ned motion predictions translate into more informed probability distributions over hypotheses and nally into a more robust tracking behavior and better occlusion handling. In experiments in indoor and outdoor environments with data from a laser range nder, the social force model leads to more accurate tracking with up to two times fewer data association errors.

[link]

Lab Meeting June 29th, 2010 (Jeff): Fully Autonomous Trajectory Estimation with Long-Range Passive RFID

Title: Fully Autonomous Trajectory Estimation with Long-Range Passive RFID

Authors: Philipp Vorst and Andreas Zell

Abstract:

We present a novel approach which enables a mobile robot to estimate its trajectory in an unknown environment with long-range passive radio-frequency identi cation
(RFID). The estimation is based only on odometry and RFID measurements. The technique requires no prior observation model and makes no assumptions on the RFID setup. In
particular, it is adaptive to the power level, the way the RFID antennas are mounted on the robot, and environmental characteristics, which have major impact on long-range RFID
measurements. Tag positions need not be known in advance, and only the arbitrary, given infrastructure of RFID tags in the environment is utilized. By a series of experiments with a
mobile robot, we show that trajectory estimation is achieved accurately and robustly.

Link:
IEEE International Conference on Robotics and Automation(ICRA), May 2010
http://www.ra.cs.uni-tuebingen.de/publikationen/2010/vorst2010icra.pdf

Lab Meeting June 8th, 2010 (Wang Li): Pictorial Structures for Object Recognition

Pictorial Structures for Object Recognition

Pedro F. Felzenszwalb
Daniel P. Huttenlocher
IJCV 61(1), 2005

Abstract
In this paper we present a computationally efficient framework for part-based modeling and recognition of objects, motivated by the pictorial structure models introduced by Fischler and Elschlager. We address the problem of using pictorial structure models to find instances of an object in an image as well as the problem of learning an object model from training examples, presenting efficient algorithms in both cases. We demonstrate the techniques
by learning models that represent faces and human bodies and using the resulting
models to locate the corresponding objects in novel images.

Paper Link

MSI Robot Vacuum

CMU PhD Thesis Defense: Search in the Physical World

Search in the Physical World

Geoffrey A. Hollinger
Carnegie Mellon University

June 01, 2010, 10:00 a.m., NSH 3305
Abstract
This thesis examines search in the physical world, which differs significantly from the searches in the digital world that we perform every day on our computers. When searching the internet, for instance, success is a matter of informed indexing that allows the information to be retrieved quickly. In these cases, there is no consideration of the physical nature of the world, and the search is not cognizant of space, time, or traversal distance. In contrast, search in the physical world must consider a target that could be continuously moving, possibly even trying to evade being found. The environment may be partially known, and the search proceeds with information gathered during the search itself. In many cases, such as guaranteeing capture of an adversarial target, the problem cannot be solved with a single searcher, and all group members must coordinate their actions with others on the team. Prior work has explored limited instances of such problems, but existing techniques either scale poorly or do not have performance guarantees.

Two of the main variations of search in the physical world are considered: efficient search and guaranteed search. During efficient search, robots move to optimize the average-case performance of the search given a model of the target’s motion. During guaranteed search, robots coordinate to provide worst-case guarantees on search time if the target is adversarial. This thesis unifies these search problems and shows them to be NP-hard, which suggests that a scalable and optimal algorithm is unlikely. Despite these hardness results, algorithms using implicit coordination can provide scalable and high-performing approximate solutions to many real-world search problems. Implicit coordination arises when robots share their locations, measurements, and/or actions to improve the plans of their teammates. In accord with this design strategy, a linearly scalable efficient search algorithm is presented that utilizes implicit coordination to achieve bounded performance. In addition, this thesis contributes a novel approach that augments the coordination with a pre-search spanning tree generation step, which leads to an anytime algorithm for guaranteed search.

With a focus on decentralized and online operation, the proposed search algorithms are extended to take into account team constraints, limited communication, and partially known environments. The techniques are illustrated using a scenario in the literature that incorporates both efficient and guaranteed search, and they are validated both in simulation and on human-robot search teams operating in the physical world. The developed framework enables teams of autonomous agents to search environments outside the scope of previous techniques, and the analysis provides insight into the complexity of multi-robot coordination problems.

[Thesis PDF]

Thesis Committee
Sanjiv Singh, Chair
Geoff Gordon
Reid Simmons
Athanasios Kehagias, Aristotle University of Thessaloniki

Department machine learning talks: Interactively Building Mashups by Demonstration

Title: Interactively Building Mashups by Demonstration
Speaker: Dr. Craig A. Knoblock, University of Southern California
Time: 10:30am, May 25 (Tue), 2010
Place: Room 210, CSIE Building

Abstract:

There are a number of tools and services available now for building mashups on the Web. However, many of the tools for constructing mashups reply on a widget paradigm, where users must select, customize, and connect widgets to build the desired application. While this approach does not require programming, the users must still understand programming concepts to successfully create a mashup. In this talk I describe our programming-by-demonstration approach to building mashups by example. Instead of requiring a user to select and customize a set of widgets, the user simply demonstrates the integration task by example. I will describe how this approach addresses the problems of extracting data from various sources, cleaning and modeling the extracted data, integrating the data across sources, and visualizing the integrated results in a geospatial context. We implemented these ideas in a system called Karma and evaluated Karma on a set of 20 users and showed that compared to other mashup construction tools, Karma allowed more of the users to successfully build mashups and made it possible to build these mashups significantly faster compared to using a widget-based approach.

This research is joint work with Shubham Gupta, Pedro Szekely, and Rattapoom Tuchinda.

Short Biography:

Dr. Craig Knoblock is a Research Professor in Computer Science and a Senior Project Leader in the Information Sciences Institute at the University of Southern California (USC). He received both his M.S. and Ph.D. in Computer Science from Carnegie Mellon and his B.S. from Syracuse University. His current research interests include information integration, information extraction, machine learning, users interfaces, constraint reasoning, geospatial data fusion, and bioinformatics. He has published one book and over 200 articles, book chapters, and conference papers on his research. He has served on the Senior Program Committees of the National Artificial Intelligence Conference, the International Joint Conference on AI, the International Semantic Web Conference, and the International Conference on Intelligent User Interfaces. He was program co-chair for the 2008 AAAI track on AI and the Web and he is conference chair for the 2011 International Joint Conference on AI (IJCAI). He is on the editorial board of Artificial Intelligence, AAAI Press, Computational Intelligence, and the Journal on Foundations and Trends in Web Science. He is a Fellow of the Association for the Advancement of Artificial Intelligence (AAAI), a Distinguished Scientist of the Association of Computing Machinery (ACM), a Trustee of the International Joint Conference on Artificial Intelligence (IJCAI), and past President of the International Conference on Automated Planning and Scheduling (ICAPS). He has started two companies, Fetch Technologies and Geosemble Technologies, based on his research at USC.

Department machine learning talks: Transfer Learning with Applications

Title: Transfer Learning with Applications
Speaker: Prof. Qiang Yang, Hong Kong University of Science and Technology
Time: 11:15am, May 25 (Tue), 2010
Place: Room 210, CSIE Building

Abstract:

Transfer learning is a new machine learning and data mining framework that allows the training and test data to come from different distributions or feature spaces. We can find many novel applications of machine learning and data mining where transfer learning is necessary. In this talk, I will give an introduction to transfer learning and then highlight some important applications such as text and image classification, sensor network data mining and activity recognition, collaborative filtering and bioinformatics. I will also discuss some potential future directions of transfer learning.

Short Biography:

Qiang Yang is a professor of the Department of Computer Science and Engineering at the Hong Kong University of Science and Technology. He is also an adjunct professor at Peking University, Beijing, and at Zhongshan University in Guangzhou, China. He received his PhD degree from the University of Maryland, College Park. His research interests include AI planning and sensor-based activity recognition, machine learning and case-based reasoning, and data mining. He is a senior member of the IEEE, the AAAI, and the ACM, and an associate editor for the IEEE Transactions on Knowledge and Data Engineering and IEEE Intelligent Systems, as well as the International Journal of Knowledge and Information Systems. More information about him can be found at http://www.cse.ust.hk/~qyang/

Lab Meeting June 8th, 2010 (fish60): Learning to Navigate Through Crowded Environments

Peter Henry, Christian Vollmer, Brian Ferris, and Dieter Fox,
Learning to Navigate Through Crowded Environments,
in Proceedings of the IEEE International Conference on Robotics and Automation (ICRA2010), Anchorage, Alaska, May 2010

Abstract—The goal of this research is to enable mobile robots to navigate through crowded environments such as indoor shopping malls, airports, or downtown side walks. The key research question addressed in this paper is how to learn planners that generate human-like motion behavior. Our approach uses inverse reinforcement learning (IRL) to learn human-like navigation behavior based on example paths. Since robots have only limited sensing, we extend existing IRL methods to the case of partially observable environments. We demonstrate the capabilities of our approach using a realistic crowd flow simulator in which we modeled multiple scenarios in crowded environments. We show that our planner learned to guide the robot along the flow of people when the environment is crowded, and along the shortest path if no people are around.

Link

News: Innovation: Teaching robots some manners

13:07 17 May 2010 by Colin Barras

Where PCs are concerned, faster is invariably better. But things aren't so clear-cut in human society. The next generation of social robots will be better loved if they adopt more human-like behaviour – even if that means losing some of their raw efficiency.

Norihiro Hagita and colleagues at the ATR laboratories in Kyoto, Japan, asked 38 volunteers to click on a PC mouse to enlarge an image. The response was programmed to be delayed by 1 to 3 seconds. As expected, an immediate response was most favoured, and participants expressed more and more dissatisfaction as the delay lengthened.

But a version of the experiment that involved a humanoid robot threw up a surprising result. The volunteers were asked to tell the robot to take out the rubbish, and the robot verbally acknowledged the request. This time an immediate response – beginning the moment the volunteer finished talking – was considered less welcome than one that was delayed by a second.

...

See the full article here.

News: Software that Learns by Watching

KarDo learns how to perform common IT support tests by observing what the experts do.
By Duncan Graham-Rowe
WEDNESDAY, MAY 19, 2010

Overworked and much in demand, IT support staff can't be in two places at once. But software designed to watch and learn as they carry out common tasks could soon help--by automatically performing the same jobs across different computers.

The new software system, called KarDo, was developed by researchers at MIT. It can automatically configure an e-mail account, install a virus scanner, or set up access to a virtual private network, says MIT's Dina Katabi, an associate professor at MIT.

Crucially, the software just needs to watch an administrator perform this task once before being able to carry out the same job on computers running different software. Businesses spend billions of dollars each year on simple and repetitive IT tasks, according to reports from the analyst groups Forrester and Gartner. KarDo could reduce these costs by as much as 20 percent, Katabi says.

See the full article here.

Lab Meeting June 1st, 2010 (Kuen-Han): Multiframe Motion Segmentation with Missing Data Using PowerFactorization and GPCA(IJCV 2008)

Title: Multiframe Motion Segmentation with Missing Data Using
PowerFactorization and GPCA
Authors: René Vidal · Roberto Tron · Richard Hartley

Abstract: We consider the problem of segmenting multiple
rigid-body motions from point correspondences in multiple
affine views. We cast this problem as a subspace clustering
problem in which point trajectories associated with each
motion live in a linear subspace of dimension two, three or
four. Our algorithm involves projecting all point trajectories
onto a 5-dimensional subspace using the SVD, the Power-
Factorization method, or RANSAC, and fitting multiple linear
subspaces representing different rigid-body motions to
the points in R5 using GPCA. Unlike previous work, our
approach does not restrict the motion subspaces to be
fourdimensional and independent. Instead, it deals gracefully
with all the spectrum of possible affine motions: from twodimensional
and partially dependent to four-dimensional and fully independent.

Our algorithm can handle the case of missing data, meaning
that point tracks do not have to be visible in all images, by
using the PowerFactorization method to project the data. In
addition, our method can handle outlying trajectories by using
RANSAC to perform the projection.

We compare our approach to other methods on a database of
167 motion sequences with full motions, independent motions,
degenerate motions, partially dependent motions, missing data,
outliers, etc. On motion sequences with complete data our
method achieves a misclassification error of less that 5% for
two motions and 29% for three motions.

paper link

Lab Meeting June 1, 2010 (Jimmy): Object Recognition in 3D Point Clouds Using Web Data and Domain Adaptation

Title: Object Recognition in 3D Point Clouds Using Web Data and Domain Adaptation
In: IJRR2010
Authors: Kevin Lai and Dieter Fox

Abstract
In recent years, object detection has become an increasingly active field of research in robotics. An important problem in object detection is the availability of a sufficient amount of labeled training data to learn good classifiers. In this paper we show how to significantly reduce the need for manually labeled training data by leveraging data sets available on the World Wide Web. Specifically, we show how to use objects from Google’s 3D Warehouse to train an object detection system for 3D point clouds collected by robots navigating through both urban and indoor environments. In order to deal with the different characteristics of the web data and the real robot data, we additionally use a small set of labeled point clouds and perform domain adaptation. Our experiments demonstrate that additional data taken from the 3D Warehouse along with our domain adaptation greatly improves the classification accuracy on real-world environments.

[pdf]

Lab Meeting May, 18 (Gary) : "2D vs. 3D Deformable Face Models: Representational Power, Construction, and Real-Time Fitting "( IJCV 2007)

Title:
2D vs. 3D Deformable Face Models: Representational Power, Construction, and Real-Time Fitting

Authors: Iain Matthews, Jing Xiao, Simon Baker


Abstract:

Model-based face analysis is a general paradigm with applications that include face recognition, expression recognition, lip-reading, head pose estimation, and gaze estimation. A face model is first constructed from a collection of training data, either 2D images or 3D range scans. The face model is then fit to the input image(s) and the model parameters used in whatever the application is. Most existing face models can be classified as either 2D (e.g. Active Appearance Models) or 3D (e.g. Morphable Models). In this paper we compare 2D and 3D face models along three axes: (1) representational power, (2) construction, and (3) real-time fitting. For each axis in turn, we outline the differences that result from using a 2D or a 3D face model.


link

ICRA 2010 Awards - Best Cognitive Robotics Paper

• Gesture-Based Human-Robot Jazz Improvisation: Guy Hoffman and Gil Weinberg (Video)
  
• Categorizing Object-Action Relations from Semantic Scene Graphs: Eren Erdal Aksoy, Alexey Abramov, Florentin Woergoetter and Babette Dellen
• Mapping Indoor Environments Based on Human Activity: Slawomir Grzonka, Frederic Dijoux, Andreas Karwath and Wolfram Burgard
• Biomimetic Motor Behavior for Simultaneous Adaptation of Force, Impedance and Trajectory in Interaction Tasks: Gowrishankar Ganesh, Alin Albu-Schäffer, Masahiko Haruno, Mitsuo Kawato and Etienne Burdet
  

ICRA 2010 Awards - KUKA Service Robotics Best Paper

• Visual Path Following on a Manifold in Unstructured Three-Dimensional Terrain: Paul Timothy Furgale and Timothy Barfoot (Video)
  
• Efficiency Based Modulation for Wheelchair Driving Collaborative Control: Christina Urdiales, Manuel Fernandez-Carmona, Jose Manuel Peula Palacios, Roberta Annicchiarico, Francisco Sandoval and Carlo Caltagirone
• Anytime Online Novelty Detection for Vehicle Safeguarding: Boris Sofman, James Bagnell and Anthony Stentz
• Pulling Open Doors and Drawers: Coordinating an Omni-Directional Base and a Compliant Arm with Equilibrium Point Control: Advait Jain and Charlie Kemp (Video)
  

ICRA 2010 Awards - Best Conference Paper

• Integrated Silicon-PDMS Process for Microbot Mechanisms: Aaron P. Gerratt, Ivan Penskiy and Sarah Bergbreiter
• The Distributed Flight Array: Raymond Oung, Frederic Bourgault, Matthew Donovan and Rafaello D'Andrea
• MoonWalker, a Lower Limb Exoskeleton able to Sustain Bodyweight using a Passive Force Balancer: Sebastien Krut, Michel Benoit, Etienne Dombre and François Pierrot
• Fast, Robust Quadruped Locomotion Over Challenging Terrain: Mrinal Kalakriskhan, Jonas Buchli, Peter Pastor, Michael Mistry and Stefan Schaal

ICRA 2010 Awards - Best Student Paper

• Non-Ideal Swimming of Artificial Bacterial Flagella Near a Surface: Kathrin Eva Peyer, Li Zhang, Bradley Kratochvil and Bradley J. Nelson
• Feature Detection and Matching in Images with Radial Distortion: Miguel Lourenço, Joăo P. Barreto and Abed Malti
• Mapping Indoor Environments Based on Human Activity: Slawomir Grzonka, Frederic Dijoux, Andreas Karwath and Wolfram Burgard
• Toward the Set of Frictional Velocity Fields Generable by 6-Degree-Of-Freedom Oscillatory Motion of a Rigid Plate: Thomas Vose, Paul Umbanhowar and Kevin Lynch
  

ICRA 2010 Awards - Best Medical Robotics Paper

• Superhuman Performance of Surgical Tasks by Robots Using Iterative Learning from Human-Guided Demonstrations: Jur van den Berg, Daniel Duckworth, Stephen Miller, Humphrey Hu, Andrew Wan, Xiao-Yu Fu, Ken Goldberg and Pieter Abbeel
• Wireless Control of Magnetic Helical Microbots Using a Rotating-Permanent-Magnet Manipulator: Thomas Fountain, Prem Kailat and Jake Abbott
• Design and Control of Motion Compensation Cardiac Catheters: Samuel Benjamin Kesner and Robert D. Howe
• Robotically Generated Force Fields for Stroke Patient Pelvic Obliquity Gait Rehabilitation: Maciej Pietrusinski, Iahn Cajigas, Yuhannes Mizikacioglu, Paolo Bonato and Constantinos Mavroidis
  

ICRA 2010 Awards - Best Automation Paper

• Continuous Path Tracing by a Cable-Suspended, Under-Actuated Robot: The Winch-Bot: Daniel Cunningham and Harry Asada
• Hybrid Dual Actuator Unit: A Design of a Variable Stiffness Actuator Based on an Adjustable Moment Arm Mechanism: Byeong-Sang Kim and Jae-Bok Song
• New Magnetic Rotational Drive by Use of Magnetic Particles with Specific Gravity Smaller Than a Liquid: Makoto Nokata, Hiromi Masuka and Satoshi Kaitamura
• Rao-Blackwellized Particle Filtering for Probing-Based 6-DOF Localization in Robotic Assembly: Yuichi Taguchi and Tim K. Marks
  

ICRA 2010 Awards - Best Vision Paper

• Real-time Monocular SLAM: Why Filter?: Hauke Strasdat, J.M.M. Montiel and Andrew J. Davison
• Projected Texture Stereo: Kurt Konolige
• Using Stereo for Object Recognition: Scott Helmer and David Lowe
• FAB-MAP 3D: Topological Mapping with Spatial and Visual Appearance: Rohan Paul and Paul Newman
  

ICRA 2010 Awards - Best Manipulation Paper

• Octo-Mag: An Electromagnetic System for 5-DOF Wireless Micromanipulation: Michael Kummer, Jake Abbott, Bradley Kratochvil, Ruedi Borer, Ali Sengul and Bradley J. Nelson
• Planning Pre-Grasp Manipulation for Transport Tasks: Lillian Chang, Siddhartha Srinivasa and Nancy S. Pollard
• Representation and Constrained Planning of Manipulation Strategies in the Context of Programming by Demonstration: Rainer Jäkel, Sven R. Schmidt-Rohr, Martin Lösch and Rüdiger Dillmann
• A Micromanipulation System for Single Cell Deposition: Zhe Lu, Chris Moraes, Yan Zhao, LiDan You, Craig A. Simmons and Yu Sun
  

ICRA 2010 Awards - Best Video

This post, along with the following posts, will provide links to the best paper awards and finalists for each category in ICRA 2010, with the best papers listed first.

• Minimalistic, Dynamic, Tube Climbing Robot: Amir Degani, Siyuan Feng, Howie Choset and Matthew T. Mason (Video)
  
• Octo-Mag: An Electromagnetic System for 5-DOF Wireless Manipulation: Bradley Kratochvil, Michael Kummer, Jake Abbott, Ruedi Borer, Olgaç Ergeneman and Bradley J. Nelson (Video)
  
• From Compliant Balancing to Dynamic Walking on Humanoid Robot: Integration of CNS and CPG: Sang-Ho Hyon (Video)
  
• RANGE - Robust Autonomous Navigation in GPS-Denied Environments: Abraham Bachrach, Anton de Winter, Ruijie He, Garrett Hemann, Samuel Prentice and Nicholas Roy (Video)
  

Lab Meeting May 11 (Nicole): Active Audition Using the Parameter-less Self-organising Map (Auton Robot 2008)

Title: Active Audition Using the Parameter-less Self-organising Map
Authors: Erik Berglund · Joaquin Sitte · Gordon Wyeth

Autonomous Robots Volume 24, Number 4, 2008/5

Abstract:
This paper presents a novel method for enabling a robot to determine the position of a sound source in three dimensions using just two microphones and interaction with its environment. The method uses the Parameter-Less Self-Organising Map (PLSOM) algorithm and Reinforcement Learning (RL) to achieve rapid, accurate response. We also introduce a method for directional filtering using the PLSOM. The presented system is compared to a similar system to evaluate its performance.

[Link]

Lab Meeting May 11 (Shao-Chen): Decentralized Localization of Sparsely-Communicating Robot Networks: A Centralized-Equivalent Approach(IEEE T-RO 2010)

Title: Decentralized Localization of Sparsely-Communicating Robot Networks: A Centralized-Equivalent Approach
Authors: Leung, K.Y.K.;   Barfoot, T.D.;   Liu, H.;   Inst. for Aerosp. Studies, Univ. of Toronto, Toronto, ON, Canada

IEEE TRANSACTIONS ON ROBOTICS, VOL. 26, NO. 1, FEBRUARY 2010

Abstract:
Finite-range sensing and communication are factors in the connectivity of a dynamic mobile-robot network. State estimation becomes a difficult problem when communication connections allowing information exchange between all robots are not guaranteed. This paper presents a decentralized state-estimation algorithm guaranteed to work in dynamic robot networks without connectivity requirements. We prove that a robot only needs to consider its own knowledge of network topology in order to produce an estimate equivalent to the centralized state estimate whenever possible while ensuring that the same can be performed by all other robots in the network. We prove certain properties of our technique and then it is validated through simulations. We present a comprehensive set of results, indicating the performance benefit in different network connectivity settings, as well as the scalability of our approach.

Link

ICRA 2010 Day 1 Report:

I will attend the Best Practice in 3D Perception and Modeling for Mobile Manipulation workshop on May 3rd. The web page of this workshop is here. The papers of this workshop are available at the lab server.

-Bob

Lab Meeting April 27th (Andi): Error Modeling and Calibration of Exteroceptive Sensors for Accurate Mapping Applications

Authors: James P. Underwood, Andrew Hill, Thierry Peynot, and Steven J. Scheding
ARC Centre of Excellence for Autonomous Systems, Australian Centre for Field Robotics

Abstract: Reliable robotic perception and planning are critical to performing autonomous actions in uncertain, unstructured environments. In field robotic systems, automation is achieved by interpreting exteroceptive sensor information to infer something about the world. This is then mapped to provide a consistent spatial context, so that actions can be planned around the predicted future interaction of the robot and the world. The whole system is as reliable as the weakest link in this chain. In this paper, the term mapping is used broadly to describe the transformation of range-based exteroceptive sensor data (such as LIDAR or stereo vision) to a fixed navigation frame, so that it can be used to form an internal representation of the environment. The coordinate transformation from the sensor frame to the navigation frame is analyzed to produce a spatial error model that
captures the dominant geometric and temporal sources of mapping error. This allows the mapping accuracy to be calculated at run time. A generic extrinsic calibration method for exteroceptive range-based sensors is then presented to determine the sensor location and orientation. This allows systematic errors in individual sensors to be minimized, and when multiple sensors are used, it minimizes the systematic contradiction between them to enable reliable multisensor data fusion. The mathematical derivations at the core of this model are not particularly novel or complicated, but the rigorous analysis and application to field robotics seems to be largely absent from the literature to date. The techniques in this paper are simple to implement, and they offer a significant improvement to the accuracy, precision, and integrity of mapped information. Consequently, they should
be employed whenever maps are formed from range-based exteroceptive sensor data.

full paper

Lab Meeting April 20th, 2010 (KuoHuei): Directing Crowd Simulations Using Navigation Fields (TVCG 2010)

Title: Directing Crowd Simulations Using Navigation Fields

(to appear in IEEE Transactions on Visualization and Computer Graphics)

Authors: Sachin Patil,Jur van den Berg,Sean Curtis,Ming Lin,Dinesh Manocha

Abstract: We present a novel approach to direct and control virtual crowds using navigation fields. Our method guides one or more agents towards desired goals based on guidance fields. The system allows the user to specify these fields by either sketching paths directly in the scene via an intuitive authoring interface or by importing motion flow fields extracted from crowd video footage. We propose a novel formulation to blend input guidance fields to create singularity-free, goal-directed navigation fields. Our method can be easily combined with most current local collision-avoidance methods and we use two such methods as examples to highlight the potential of our approach. We illustrate its performance on several simulation scenarios.

Link: [web] [pdf]

Lab Meeting April 20th, 2010 (Jeff): iSAM: Incremental Smoothing and Mapping

Title: iSAM: Incremental Smoothing and Mapping

Authors: Michael Kaess, Ananth Ranganathan, and Frank Dellaert

Abstract:

In this paper, we present incremental smoothing and mapping (iSAM), which is a novel approach to the simultaneous localization and mapping problem that is based on fast incremental matrix factorization. iSAM provides an efficient and exact solution by updating a QR factorization of the naturally sparse smoothing information matrix, thereby recalculating only those matrix entries
that actually change. iSAM is efficient even for robot trajectories with many loops as it avoids unnecessary fill-in in the factor matrix by periodic variable reordering. Also, to enable data association in real time, we provide efficient algorithms to access the estimation uncertainties of interest based on the factored information matrix. We systematically evaluate the different components of iSAM as well as the overall algorithm using various simulated and realworld
datasets for both landmark and pose-only settings.

Link:
IEEE Transactions on Robotics, Vol. 24, No. 6, December 2008:1365-1378
http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=4682731

News: Grand Challenges of Science: Robotics

In January, DISCOVER and the National Science Foundation continued their Grand Challenges event series with a panel discussion at Carnegie Mellon University exploring the dynamic world of robotics.

The panel included four eminent roboticists—Javier Movellan from University of California San Diego, Rodney Brooks from the Massachussetts Institute of Technology, William "Red" Whittaker from Carnegie Mellon University, and Robyn Murphy from Texas A&M University—who discussed some the big questions on the future of their field: How will robots transform industry, health care, and warfare? Will they ever be our equals? The conversation was moderated by DISCOVER editor-in-chief Corey Powell.

See the video

Lab Meeting April 13, 2010 (Alan) - Information-Based Compact Pose SLAM (T-RO 2010)

Title: Information-Based Compact Pose SLAM

Authors: Viorela Ila, Josep M. Porta, and Juan Andrade-Cetto, Member, IEEE

IEEE TRANSACTIONS ON ROBOTICS, VOL. 26, NO. 1, FEBRUARY 2010

Abstract—Pose SLAMis the variant of simultaneous localization and map building (SLAM) is the variant of SLAM, in which only the robot trajectory is estimated and where landmarks are only used to produce relative constraints between robot poses. To reduce the computational cost of the information filter form of PoseSLAM and, at the same time, to delay inconsistency as much as possible, we introduce an approach that takes into account only highly informative loop-closure links and nonredundant poses. This approach includes constant time procedures to compute the distance between poses, the expected information gain for each potential link, and the exact marginal covariances while moving in open loop, as well as a procedure to recover the state after a loop closure that, in practical situations, scales linearly in terms of both time and memory. Using these procedures, the robot operates most of the time in open loop, and the cost of the loop closure is amortized over long trajectories. This way, the computational bottleneck shifts to data association, which is the search over the set of previously visited poses to determine good candidates for sensor registration. To speed up data association, we introduce a method to search for neighboring poses whose complexity ranges from logarithmic in the usual case to linear in degenerate situations. The method is based on organizing the pose information in a balanced tree whose internal levels are defined using interval arithmetic. The proposed Pose-SLAM approach is validated through simulations, real mapping sessions, and experiments using standard SLAM data sets.

Link

Researchers develop a robot that folds towels

More than a household convenience, the project is a breakthrough in the robotic manipulation of non-rigid objects

Check out their ICRA 2010 paper.

Lab Meeting 3/29, 2010 (swem): MonoSLAM: Real-Time Single Camera SLAM

MonoSLAM: Real-Time Single Camera SLAM

Andrew J. Davison, Ian D. Reid, Member, IEEE, Nicholas D. Molton, and Olivier Stasse, Member, IEEE

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, VOL. 29, NO. 6, JUNE 2007

[link]

Abstract—We present a real-time algorithm which can recover the 3D trajectory of a monocular camera, moving rapidly through a

previously unknown scene. Our system, which we dub MonoSLAM, is the first successful application of the SLAM methodology from

mobile robotics to the “pure vision” domain of a single uncontrolled camera, achieving real time but drift-free performance inaccessible

to Structure from Motion approaches. The core of the approach is the online creation of a sparse but persistent map of natural

landmarks within a probabilistic framework. Our key novel contributions include an active approach to mapping and measurement, the

use of a general motion model for smooth camera movement, and solutions for monocular feature initialization and feature orientation

estimation. Together, these add up to an extremely efficient and robust algorithm which runs at 30 Hz with standard PC and camera

hardware. This work extends the range of robotic systems in which SLAM can be usefully applied, but also opens up new areas. We

present applications of MonoSLAM to real-time 3D localization and mapping for a high-performance full-size humanoid robot and live

augmented reality with a hand-held camera.

NTU talk: Image-Based Mobile Robot Navigation Design

Title: Image-Based Mobile Robot Navigation Design
Speaker: Prof. Kai-Tai Song, National Chiao Tung University
Time: 2:20pm, April 2 (Fri), 2010
Place: Room 103, CSIE Building


Abstract: Autonomous navigation is a basic function of every mobile robot. It is important for a mobile robot to travel and arrive at a desired location autonomously. In order to accomplish this function, the robot needs to observe its surroundings and locate itself through acquired environmental information. Various sensors have been utilized for acquiring information from environment. Among these perception sensors, image-based systems are most promising for practical application in daily-life scenarios. In this talk, I will first give a general perspective of domestic and service robots today. Then the design issues of mobile robot navigation will be discussed. Two approaches to visual navigation design will be presented. One employs an omni-directional camera, the other uses a normal webcam. In this presentation, the design and implementation of the navigation system will be discussed. Some interesting video clips of autonomous mobile robots will be presented.

Short Biography: Kai-Tai Song received his Ph.D. degree in mechanical engineering from Katholieke Universiteit Leuven, Belgium in 1989. Since 1989 he has been a faculty member and is currently a Professor with National Chiao Tung University (NCTU). From 2007 to 2009, he served as the Associate Dean of the R & D Office of NCTU. He is currently the Director of Institute of Electrical and Control Engineering of NCTU. He has served as the Chairman of IEEE Robotics & Automation Chapter, Taipei Section in the term of 1999. His is a directorate of Robotics Association Taiwan, Taiwan Robotics Society, Taiwan Association of System Science and Engineering, and Chinese Automatic Control Society.
His areas of research interest include mobile robotics, image processing, visual tracking, human-robot interaction, and mechatronics.

Lab Meeting 3/23, 2010 (fish60): Learning to Search: Functional Gradient Techniques for Imitation Learning

I will try to present this one:

Learning to Search: Functional Gradient Techniques for Imitation Learning
Nathan Ratliff, David Silver, J. Andrew Bagnell

Submitted to Autonomous Robotics Special Issue on Robot Learning, 2009

[download draft]

Abstract:
While planning algorithms have shown success in many real-world applications ranging from legged locomotion to outdoor unstructured navigation, such algorithms rely on fully specified cost functions that map sensor readings and environment models to quantifiable costs. Such cost functions are usually manually designed and programmed. Recently, a set of techniques has been developed that explore learning these functions from expert human demonstration. These algorithms apply an inverse optimal control approach to find a cost function for which planned behavior mimics an expert's demonstration.
The work we present extends the Maximum Margin Planning (MMP) frame- work to admit learning of more powerful, non-linear cost functions. These algorithms, known collectively as LEARCH (LEArning to seaRCH ), are simpler to implement than most existing methods, more efficient than previous attempts at non-linearization, more naturally satisfy common constraints on the cost function, and better represent our prior beliefs about the function's form.

NTU talk: 3D Reconstruction from Images

Title: 3D Reconstruction from Images

Shang-Hong Lai
Department of Computer Science
National Tsing Hua University

Time: 2:20pm, March 26 (Fri), 2010
Place: Room 103, CSIE Building

Abstract: The goal of 3D computer vision is to recover real-world three-dimensional information of the scene or objects from 2D images. There have been many different approaches with various image acquisition set-ups to achieve the 3D reconstruction in computer vision. In this talk, I will discuss some 3D reconstruction systems and some researches on 3D reconstruction in my lab. I will present our researches on 3D reconstruction from multi-view images, two-view (stereo) images, and a single image. Some 3D reconstruction results will be shown to demonstrate the performance of our systems.

Bio: Shang-Hong Lai received the BS and MS degrees in electrical engineering from National Tsing Hua University in 1986 and 1988, respectively. He also received the PhD degree from University of Florida, Gainesville, USA, in 1995. He worked for Siemens Corporate Research in Princeton, USA, as a research scientist from 1995 to 2000. Then, Dr. Lai returned to Taiwan to join department of computer science, National Tsing Hua University, as a faculty member. He is currently a professor and associate chair in the same department.

CMU PhD Thesis proposal: Data-driven Scene Parsing With the Visual Memex

Data-driven Scene Parsing With the Visual Memex

Tomasz Malisiewicz

Carnegie Mellon University

February 18, 2010, 4:00 p.m., NSH 3305

Abstract: This proposal is concerned with the problem of image understanding. Given a single static image, the goal is to explain the entire image by recognizing all of the objects depicted in the image. We formulate the problem of image understanding as image parsing -- breaking up the image into semantically meaningful regions and recognizing the objects embedded in each region. In our approach we strive to obtain a dense understanding of the image by not leaving any portion of the image unexplained. While most approaches to scene understanding formulate the problem as that of recognizing abstract object categories (and for object asking “what is this?”), we use a data-driven model of recognition more akin to memory (and ask the question: “what is this like?”). We present an exemplar-based framework for reasoning about objects and their relationships in images dubbed the Visual Memex. The Visual Memex is a non-parametric graph-based model of objects which encodes two types of object relationships: visual similarity between object exemplars, and 2D spatial context between objects in a single image. We use a region-based representation of exemplar objects which has been shown to be superior to the popular rectangular window approach for a wide array of things and stuff found in natural scenes. During training, we learn a set of similarity functions per-exemplar and formulate recognition as association between automatically extracted regions from the input image and exemplar regions in the Visual Memex. We use both bottom-up image segmentation, mid-level reasoning about segment relationships as well as spatial relationships between exemplars in the Visual Memex as complementary sources of object hypotheses. I propose an iterative image parsing framework which builds an interpretation of an input image by iteratively conditioning on a current (partial) interpretation and generating novel segment hypotheses using low-level, mid-level, and high-level cues. An evaluation is proposed which evaluates the system with respect to recognition as well as segmentation on real world scenes from LabelMe.

Thesis Committee
Alexei A. Efros, Chair
Martial Hebert
Takeo Kanade
Pietro Perona, California Institute of Technology

CMU talk: Simon J.D. Prince, Monday, Feb 8, NSH 1507, 3pm-4pm

Title: Modeling Facial Images with Patches
Speaker: Simon J.D. Prince

Abstract:

Faces are one of the most studied object classes in computer vision. Performance is very good for tasks such as identity recognition and gender classification when the pose, lighting and expression are controlled. However, in uncontrolled conditions, these tasks remain challenging. Part of the reason for this limitation is the choice of representation: for example, faces have variously been modeled as subspaces and constellations of features, but these representations have only a limited ability to describe uncontrolled facial images. In this talk, I will present several experiments in which we have investigated representing faces with a regular grid of patches. This type of model can better capture the complex multimodal appearance of uncontrolled faces. I will present models for both gender recognition (or more generally classification of facial characteristic) and pose estimation (regression). I will also show how to extend these patch-based models to allow generation of near photo-realistic images of novel faces.

Bio:

Simon Prince was an undergraduate at UCL where he studied Psychology. His doctoral work was at the University of Oxford, in the Department of Experimental Psychology where he investigated human stereo vision using psychophysics. He subsequently worked in the Laboratory of Physiology in Oxford for two years as a post-doc with Andrew Parker studying stereo vision using single unit electro-physiology. In 2001 he became a post-doctoral research fellow in the Department of Electrical and Computer Engineering in the National University of Singapore working on augmented reality. Following this, he moved to Toronto, Canada, where he worked as a post-doc in computer vision for James Elder in the Centre for Vision Research in York University. Since 2005 he has been a faculty member in the department of computer science at University College London. His current interests include image segmentation, face recognition, optical tomography and object recognition.

Lab Meeting January 19, 2010 (Jimmy): Semi-supervised Learning in Gigantic Image Collections

Title: Semi-supervised Learning in Gigantic Image Collections
Authors: Rob Fergus, Yair Weiss, and Antonio Torralba
In : NIPS2009

Abstract :
With the advent of the Internet it is now possible to collect hundreds of millions of images. These images come with varying degrees of label information. “Clean labels” can be manually obtained on a small fraction, “noisy labels” may be extracted automatically from surrounding text, while for most images there are no labels at all. Semi-supervised learning is a principled framework for combining these different label sources. However, it scales polynomially with the number of images, making it impractical for use on gigantic collections with hundreds of millions of images and thousands of classes. In this paper we show how to utilize recent results in machine learning to obtain highly efficient approximations for semi-supervised learning that are linear in the number of images. Specifically, we use the convergence of the eigenvectors of the normalized graph Laplacian to eigenfunctions of weighted Laplace-Beltrami operators. Our algorithm enables us to apply semi-supervised learning to a database of 80 million images gathered from the Internet.

[paper][slides]

Lab Meeting January 19, 2010(Chung-Han):A collection of outdoor robotic datasets with centimeter-accuracy ground truth

Title : A collection of outdoor robotic datasets with centimeter-accuracy ground truth
Author : Jose-Luis Blanco, Francisco-Angel Moreno, Javier-Gonzalez
In : Autonomous Robots (2009)

Abstract :
The lack of publicly accessible datasets with a reliable ground truth has prevented in the past a fair and coherent comparison of different methods proposed in the mobile robot Simultaneous Localization and Mapping (SLAM) literature. Providing such a ground truth becomes specially challenging in the case of visual SLAM, where the world model is 3-dimensional and the robot path is 6-dimensional. This work addresses both the practical and theoretical issues found while building a collection of six outdoor datasets. It is discussed how to estimate the 6-d vehicle path from readings of a set of three Real Time Kinematics (RTK) GPS receivers, as well as the associated uncertainty bounds that can be employed to evaluate the performance of SLAM methods. The vehicle was also equipped with several laser scanners, from which reference point clouds are built as a testbed for other algorithms such as segmentation or surface fitting. All the datasets, calibration information and associated software tools are available for download http://babel.isa.uma.es/mrpt/papers/dataset2009/.

[Full Text]

CMU PhD Thesis Defense: Mapping Large Urban Environments with GPS-Aided SLAM

CMU RI PhD Thesis Defense

Justin Carlson
Mapping Large Urban Environments with GPS-Aided SLAM
December 03, 2009, 9:00 a.m., GHC 6115

Abstract

Simultaneous Localization and Mapping (SLAM) has been an active area of research for several decades, and has become a foundation of indoor mobile robotics. However, although the scale and quality of results have improved markedly in that time period, no current technique can effectively handle city-sized urban areas.

The Global Positioning System (GPS) is an extraordinarily useful source of localization information. Unfortunately, the noise characteristics of the system are complex, arising from a large number of sources, some of which have large autocorrelation. Incorporation of GPS signals into SLAM algorithms requires using low-level system information and explicit models of the underlying system to make appropriate use of the information. The potential benefits of combining GPS and SLAM include increased robustness, increased scalability, and improved accuracy of localization.

This dissertation presents a theoretical background for GPS-SLAM fusion. The presented model balances ease of implementation with correct handling of the highly colored sources of noise in a GPS system.. This utility of the theory is explored and validated in the framework of a simulated Extended Kalman Filter driven by real-world noise.

The model is then extended to Smoothing and Mapping (SAM), which overcomes the linearization and algorithmic complexity limitations of the EKF formulation. This GPS-SAM model is used to generate a probabilistic landmark-based urban map covering an area an order of magnitude larger than previous work.

[LINK][PDF]

Thesis Committee
Charles Thorpe, Chair
Brett Browning
Martial Hebert
Frank Dellaert, Georgia Institute of Technology

Lab Meeting January 6th, 2010 (Any): Learning General Optical Flow Subspaces for Egomotion Estimation and Detection of Motion Anomalies

Author: Richard Roberts, Christian Potthast and Frank Dellaert

IEEE Computer Society Conference on Computer Vision and Pattern Recognition 2009 (CVPR'09)

Abstract--This paper deals with estimation of dense optical flow and ego-motion in a generalized imaging system by exploiting probabilistic linear subspace constraints on the flow. We deal with the extended motion of the imaging system through an environment that we assume to have some degree of statistical regularity. For example, in autonomous ground vehicles the structure of the environment around the vehicle is far from arbitrary, and the depth at each pixel is often approximately constant. The subspace constraints hold not only for perspective cameras, but in fact for a very general class of imaging systems, including catadioptric and multiple-view systems. Using minimal assumptions about the imaging system, we learn a probabilistic subspace constraint that captures the statistical regularity of the scene geometry relative to an imaging system. We propose an extension to probabilistic PCA (Tipping and Bishop, 1999) as a way to robustly learn this subspace from recorded imagery, and demonstrate its use in conjunction with a sparse optical flow algorithm. To deal with the sparseness of the input flow, we use a generative model to estimate the subspace using only the observed flow measurements. Additionally, to identify and cope with image regions that violate subspace constraints, such as moving objects, objects that violate the depth regularity, or gross flow estimation errors, we employ a per-pixel Gaussian mixture outlier process. We demonstrate results of finding the optical flow subspaces and employing them to estimate dense flow and to recover camera motion for a variety of imaging systems in several different environments.

Local copy

Lab Meeting January 6th, 2010 (Kuen-Han): Adaptive Relative Bundle Adjustment(RSS 2009)

Title: Adaptive Relative Bundle Adjustment(RSS 2009)
Authors: Gabe Sibley, Christopher Mei, Ian Reid, Paul Newman

Abstract

It is well known that bundle adjustment is the
optimal non-linear least-squares formulation of the simultaneous
localization and mapping problem, in that its maximum
likelihood form matches the definition of the Cramer Rao
Lower Bound. Unfortunately, computing the ML solution is
often prohibitively expensive – this is especially true during loop
closures, which often necessitate adjusting all parameters in a
loop. In this paper we note that it is precisely the choice of a single
privileged coordinate frame that makes bundle adjustment costly,
and that this expense can be avoided by adopting a completely
relative approach. We derive a new relative bundle adjustment,
which instead of optimizing in a single Euclidean space, works
in a metric-space defined by a connected Riemannian manifold.
Using an adaptive optimization strategy, we show experimentally
that it is possible to solve for the full ML solution incrementally
in constant time – even at loop closure. Our system also operates
online in real-time using stereo data, with fast appearance-based
loop closure detection. We show results for sequences of 23k
frames over 1.08km that indicate the accuracy of the approach.

paper link
technical paper

Lab Meeting December 30th, 2009 (Nicole) : Estimation of Sound Source Number and Directions under a Multi-source Environment (IROS 2009)

Title: Estimation of Sound Source Number and Directions under a Multi-source Environment (IROS 2009)

Authors: Jwu-Sheng Hu, Member IEEE, Chia-Hsing Yang, Student Member IEEE, and Cheng-Kang Wang

Abstract:

Sound source localization is an important featurein robot audition. This work proposes a sound source numberand directions estimation method by using the delayinformation of microphone array. An eigenstructure-basedgeneralized cross correlation method is proposed to estimatetime delay between microphones. Upon obtaining the time delayinformation, the sound source direction and velocity can beestimated by least square method. In multiple sound source case,the time delay combination among microphones is arrangedsuch that the estimated sound speed value falls within anacceptable range. By accumulating the estimation results ofsound source direction and using adaptive K-means++algorithm, the sound source number and directions can beestimated.

[link]

Measuring the Accuracy of Distributed Algorithms on Multi-Robot Systems

Measuring the Accuracy of Distributed Algorithms on Multi-Robot Systems

James McLurkin (UW CSE postdoc, MIT) 
October 9, 2008, 3:30 pm
EE-105

Abstract
Distributed algorithms running on multi-robot systems rely on ad-hoc networks to relay messages throughout the group. The propagation speed of these messages is large, but not infinite, and problems in algorithm execution can arise when the robot speed is a large fraction of the message propagation speed. This implies a robot speed limit, as any robot moving away from a message source faster than the message speed will never receive new information, and no algorithm can function properly on it. In this work, we focus on measuring the accuracy of multi-robot distributed algorithms. We define the Robot Speed Ratio (RSR) as the ratio of robot speed to message speed. We express it in a form that is platform-independent and captures the relationship between communications usage, robot mobility, and algorithm accuracy. We show that trade-offs between these key quantities can be balanced at design time. Finally, we present results from experiments with 50 robots that characterize the accuracy of preexisting distributed algorithms for network communication, navigation, boundary detection, and dynamic task assignment. In all cases, accuracy degrades as speed increases or communication bandwidth is reduced. In our experiments, a RSR of 0.005 allows good accuracy in all algorithms, a RSR of 0.02 allows reasonable accuracy in simple algorithms, and all algorithms tested are essentially useless at a RSR of 0.10 or higher.

[link]

Lab Meeting December 30th, 2009(Gary) : Pose Robust Face Tracking by Combining Active Appearance Models and Cylinder Head Models

Title: Pose Robust Face Tracking by Combining Active Appearance Models and Cylinder Head Models (IJCV 2008)

Author : Jaewon Sung , Takeo Kanade , Daijin Kim

Abstract:

The active appearance models (AAMs) provide
the detailed descriptive parameters that are useful for various
autonomous face analysis problems. However, they are
not suitable for robust face tracking across large pose variation
for the following reasons. First, they are suitable for
tracking the local movements of facial features within a limited
pose variation. Second, they use gradient-based optimization
techniques for model fitting and the fitting performance
is thus very sensitive to initial model parameters.
Third, when their fitting is failed, it is difficult to obtain
appropriate model parameters to re-initialize them. To alleviate
these problems, we propose to combine the active
appearance models and the cylinder head models (CHMs),
where the global head motion parameters obtained from the
CHMs are used as the cues of the AAM parameters for a
good fitting or re-initialization. The good AAM parameters
for robust face tracking are computed in the following manner.
First, we estimate the global motion parameters by the
CHM fitting algorithm. Second, we project the previously
fitted 2D shape points onto the 3D cylinder surface inversely
Third, we transform the inversely projected shape points by
the estimated global motion parameters. Fourth, we project
the transformed 3D points onto the input image and computed
the AAM parameters from them. Finally, we treat the
computed AAM parameters as the initial parameters for the
fitting. Experimental results showed that face tracking combining
AAMs and CHMs is more pose robust than that of
AAMs in terms of 170% higher tracking rate and the 115%
wider pose coverage.



link

NTU talk: Human Action Recognition Using Bag of Video Words

Title: Human Action Recognition Using Bag of Video Words
Speaker: Dr. Mubarak Shah, Agere Chair Professor of Computer Science, University of Central Florida
Time: 4:00pm, Dec 24 (Thu), 2009
Place: Room 210, CSIE building

Abstract:

The traditional approach for video analysis involves detection of objects, followed by tracking of objects from frame to frame and finally analysis of tracks for human action recognition. However, in some videos of complex scenes it is not possible to reliably detect and track objects. Therefore, recently in computer vision there has been lots of interest in the bag of video words approach, which bypasses the object detection and tracking steps. In bag of video words approach an action is described by a distribution of spatiotemporal cuboids (3D interest points).

In this talk, first I will describe a method to automatically discover the optimal number of video words clusters by utilizing the Maximization of Mutual Information (MMI). Unlike the k-means algorithm which is typically used to cluster spatiotemporal cuboids into video words based on their appearance similarity, MMI clustering further groups the video-words, such that the semantically similar video-words, e.g. words corresponding to the same part of the body during an action, are grouped in the same cluster.

The above method for human action recognition uses only one kind of features, spatiotemporal cuboids. However, single feature based representation for human action is not sufficient to capture the imaging variations (view-point, illumination etc.) and attributes of individuals (size, age, gender etc.).

Next I will present a method which uses two types of features: i) a quantized vocabulary of local spatio-temporal (ST) volumes (or cuboids), and ii) a quantized vocabulary of spin-images. To optimally combine these features, we treat different features and videos as nodes in a graph, where weighted edges between the nodes represent the strength of the relationship between entities. The graph is then embedded into a k-dimensional space subject to the criteria that similar nodes have Euclidian coordinates which are closer to each other. This is achieved by converting this constraint into a minimization problem whose solution is the eigenvectors of the graph Laplacian matrix. This procedure is known as Fiedler Embedding.

Short Biography:
Dr. Mubarak Shah, Agere Chair Professor of Computer Science, is the founding director of the Computer Visions Lab at UCF. He is a co-author of three books (Motion-Based Recognition (1997), Video Registration (2003), and Automated Multi-Camera Surveillance: Algorithms and Practice (2008)), all by Springer. He has published ten book chapters, seventy five journal and one hundred seventy conference papers on topics related to visual surveillance, tracking, human activity and action recognition, object detection and categorization, shape from shading, geo registration, photo realistic synthesis, visual crowd analysis, bio medical imaging, etc.

Dr. Shah is a fellow of IEEE, IAPR and SPIE. In 2006, he was awarded a Pegasus Professor award, the highest award at UCF, given to a faculty member who has made a significant impact on the university, has made an extraordinary contribution to the university community, and has demonstrated excellence in teaching, research and service. He is a Distinguished ACM Speaker. He was an IEEE Distinguished Visitor speaker for 1997-2000, and received IEEE Outstanding Engineering Educator Award in 1997. He received the Harris Corporation's Engineering Achievement Award in 1999, the TOKTEN awards from UNDP in 1995, 1997, and 2000; Teaching Incentive Program awards in 1995 and 2003, Research Incentive Award in 2003, Millionaires' Club awards in 2005 and 2006, University Distinguished Researcher award in 2007, SANA award in 2007, an honorable mention for the ICCV 2005 Where Am I? Challenge Problem, and was nominated for the best paper award in ACM Multimedia Conference in 2005. He is an editor of international book series on Video Computing; editor in chief of Machine Vision and Applications journal, and an associate editor of ACM Computing Surveys journal. He was an associate editor of the IEEE Transactions on PAMI, and a guest editor of the special issue of International Journal of Computer Vision on Video Computing. He is the program co-chair of IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2008.

Lab Meeting Dez. 23rd 09 (Andi): Shape-based Recognition of 3D Point Clouds in Urban Environments

Authors: Aleksey Golovinskiy, Vladimir G. Kim, Thomas Funkhouser

International Conference on Computer Vision (ICCV), September 2009

This paper investigates the design of a system for recognizing objects in 3D point clouds of urban environments. The system is decomposed into four steps: locating, segmenting, characterizing, and classifying clusters of 3D points. Specifically, we first cluster nearby points to form a set of potential object locations (with hierarchical clustering). Then, we segment points near those locations into foreground and background sets (with a graph-cut algorithm). Next, we build a feature vector for each point cluster (based on both its shape and its context). Finally, we label the feature vectors using a classifier trained on a set of manually labeled objects. The paper presents several alternative methods for each step. We quantitatively evaluate the system and tradeoffs of different alternatives in a truthed part of a scan of Ottawa that contains approximately 100 million points and 1000 objects of interest. Then, we use this truth data as a training set to recognize objects amidst approximately 1 billion points of the remainder of the Ottawa scan.

full Paper

also:
Min-Cut Based Segmentation of Point Clouds
Aleksey Golovinskiy and Thomas Funkhouser
IEEE Workshop on Search in 3D and Video (S3DV) at ICCV, September 2009, Kyoto

Lab Meeting December 23rd, 2009 (Shao-Chen): Multi-robot SLAM with Unknown Initial Correspondence: The Robot Rendezvous Case

Title: Multi-robot SLAM with Unknown Initial Correspondence: The Robot Rendezvous Case (IROS 2006)

Authors: Xun S. Zhou and Stergios I. Roumeliotis

Abstract:

This paper presents a new approach to the multi-

robot  map-alignment  problem  that  enables  teams  of  robots  to

build joint maps without initial knowledge of their relative poses.

The  key  contribution  of  this  work  is  an  optimal  algorithm  for

merging  (not  necessarily  overlapping)  maps  that  are  created

by  different  robots  independently.  Relative  pose  measurements

between pairs of robots are processed to compute the coordinate

transformation  between  any  two  maps.  Noise  in  the  robot-

to-robot  observations,  propagated  through  the  map-alignment

process,  increases  the  error  in  the  position  estimates  of  the

transformed  landmarks,  and  reduces  the  overall  accuracy  of

the merged map. When there is overlap between the two maps,

landmarks that appear twice provide additional information, in

the form of constraints, which increases the alignment accuracy.

Landmark  duplicates  are  identified  through  a  fast  nearest-

neighbor  matching  algorithm.  In  order  to  reduce  the  compu-

tational  complexity  of  this  search  process,  a  kd-tree  is  used

to  represent  the  landmarks  in  the  original  map.  The  criterion

employed for matching any two landmarks is the Mahalanobis

distance.  As  a  means  of  validation,  we  present  experimental

results obtained from two robots mapping an area of 4,800 m

2.

[Link]

PhD Thesis Defense: Rhythmic Human-Robot Social Interaction

Marek P. Michalowski
Carnegie Mellon University
December 21, 2009, 10:00 a.m., NSH 3305

Abstract

Social scientists have identified and begun to describe rhythmic and synchronous properties of human social interaction. However, social interactions with robots are often stilted due to temporal mismatch between the behaviors, both verbal and nonverbal, of the interacting partners. This thesis brings the theory of interactional synchrony to bear on the design of social robots with a proposed architecture for rhythmic intelligence. We have developed technology that allows a robot to perceive social rhythms and to behave rhythmically. We have facilitated constrained social interactions, and designed experimental protocols, in which a robot variably synchronizes to human and/or environmental rhythms -- first in a dance-oriented task, and second in a cooperative video game. We have analyzed these interactions to understand the effects of Keepon's rhythmic attention on human performance. This thesis demonstrates that variations in a robot's rhythmic behavior have measurable effect on human rhythmic behavior and on performance in rhythmic tasks. Furthermore, human participants were able to assume and transition between the roles of leader or follower in these tasks.

Thesis Committee

Reid Simmons, Chair
Illah Nourbakhsh
Jodi Forlizzi
Hideki Kozima, Miyagi University, Japan

[link] [thesis draft]

Lab Meeting December 16th, 2009 (Casey): Monocular Vision SLAM for INdoor Aerial Vehicles

Title: Monocular Vision SLAM for INdoor Aerial Vehicles (IROS 2009)

Authors: Koray Celik, Soon-Jo Chung, Matthew Clausman, and Arun K. Somani

Abstract:

This paper presents a novel indoor navigation and ranging strategy by using a monocular camera. The proposed algorithms are integrated with simultaneous localization and mapping(SLAM) with a focus on indoor aerial vehicle applications. We experimentally validate the proposed algorithms by using a fully self-contained micro aerial vehicle (MAV) with on-board image processing and SLAM capabilities. The range measurement strategy is inspired by the key adaptive mechanisms for depth perception and pattern recognition found in humans and intelligent animals. The navigation strategy assumes an unknown, GPS-denied environment, which is representable via corner-like feature points and straight architectural lines. Experimental results show that the system is only limited by the capabilities of the camera and the availability of good corners.

[Link]

Lab Meeting December 16th, 2009 (Jeff): On measuring the accuracy of SLAM algorithms

Title: On measuring the accuracy of SLAM algorithms

Authors: Rainer Kümmerle, Bastian Steder, Christian Dornhege, Michael Ruhnke, Giorgio Grisetti, Cyrill Stachniss and Alexander Kleiner

Abstract:

In this paper, we address the problem of creating an objective benchmark for evaluating SLAM approaches. We propose a framework for analyzing the results of a SLAM approach based on a metric for measuring the error of the corrected trajectory. This metric uses only relative relations between poses and does not rely on a global reference frame. This overcomes serious shortcomings of approaches using a global reference frame to compute the error. Our method furthermore allows us to compare SLAM approaches that use different estimation techniques or different sensor modalities since all computations are made based on the corrected trajectory of the robot.
We provide sets of relative relations needed to compute our metric for an extensive set of datasets frequently used in the robotics community. The relations have been obtained by manually matching laser-range observations to avoid the errors caused by matching algorithms. Our benchmark framework allows the user to easily analyze and objectively compare different SLAM approaches.

Link:
Auton Robot 2009 27:387-407
http://www.springerlink.com/content/5u7458rl080216vr/fulltext.pdf

CMU Talk: Corridor View: Making Indoor Life Easier with Large Image Database

CMU VASC Seminar
Monday, Dec 7, 2009
1:30pm-2:30pm
NSH 1507

Corridor View: Making Indoor Life Easier with Large Image Database
Hongwen "Henry" Kang
Ph.D. Student, Robotics

Abstract:

Indoor environment poses substantial challenges for Computer Vision algorithms, due to the combined patterns that are either highly repetitive (e.g. doors), textureless (e.g. white walls), or temporally changing (e.g. posters, pedestrians). The fundamental challenge we want to tackle is the robust image matching. We proposed two approaches to address this problem, one is an iterative algorithm that combines global/local weighting strategies under bag-of-features model, the other data-mines distinctive feature vectors and uses high dimensional features directly for image matching, without quantization. Both of the approaches demonstrate significant improvements compared to straightforward image retrieval approaches, in highly confusing indoor environment. The proposed image matching techniques have broad applications. We selectively demonstrate two of them for this talk, specifically for vision impaired users living in the office environments. One application is data-driven zoomin; the other application is image composition for object pop-out.