Lab Meeting November 29, 2010 (Wang Li): Adaptive Pose Priors for Pictorial Structures (CVPR 2010)

Adaptive Pose Priors for Pictorial Structures

Benjamin Sapp
Chris Jordan
Ben Taskar

Abstract

The structure and parameterization of a pictorial structure model is often restricted by assuming tree dependency structure and unimodal, data-independent pairwise interactions, which fail to capture important patterns in the data. On the other hand, local methods such as kernel density estimation provide nonparametric flexibility but require large amounts of data to generalize well. We propose a simple semi-parametric approach that combines the tractability of pictorial structure inference with the flexibility of non-parametric methods by expressing a subset of model parameters as kernel regression estimates from a learned sparse set of exemplars. This yields query-specific, image-dependent pose priors. We develop an effective shape-based kernel for upper-body pose similarity and propose a leave-one-out loss function for learning a sparse subset of exemplars for kernel regression. We apply our techniques to two challenging datasets of human figure parsing and advance the state-of-the-art (from 80% to 86% on the Buffy dataset), while using only 15% of the training data as exemplars.

Paper Link

Lab Meeting November 29th, 2010 (Jeff): Sub-Meter Indoor Localization in Unmodified Environments with Inexpensive Sensors

Title: Sub-Meter Indoor Localization in Unmodified Environments with Inexpensive Sensors

Authors: Morgan Quigley, David Stavens, Adam Coates, and Sebastian Thrun

Abstract:

The interpretation of uncertain sensor streams for localization is usually considered in the context of a robot. Increasingly, however, portable consumer electronic devices, such as smartphones, are equipped with sensors including WiFi radios, cameras, and inertial measurement units (IMUs). Many tasks typically associated with robots, such as localization, would be valuable to perform on such devices. In this paper, we present an approach for indoor localization exclusively using the low-cost sensors typically found on smartphones. Environment modification is not needed. We rigorously evaluate our method using ground truth acquired using a laser range scanner. Our evaluation includes overall accuracy and a comparison of the contribution of individual sensors. We find experimentally that fusion of multiple sensor modalities is necessary for optimal performance and demonstrate sub-meter localization accuracy.

Link:
IEEE/RSJ International Conference on Intelligent Robots and Systems(IROS), October 2010
http://www-cs.stanford.edu/people/dstavens/iros10/quigley_etal_iros10.pdf
or
local_copy

Video:
http://www.cs.stanford.edu/people/dstavens/iros10/quigley_etal_iros10.mp4

Lab Meeting November 22, 2010 (Andi): Three-Dimensional Mapping with Time-of-Flight Cameras

Title: Three-Dimensional Mapping with Time-of-Flight Cameras

Authors: Stefan May, David Droeschel, Dirk Holz, Stefan Fuchs, Ezio Malis, Andreas Nuechter and Joachim Hertzberg

Journal of Field Robotics 2009

Abstract: This article investigates the use of time-of-flight (ToF) cameras in mapping tasks for autonomous mobile robots, in particular in simultaneous localization and mapping (SLAM) tasks. Although ToF cameras are in principle an attractive type of sensor for threedimensional (3D) mapping owing to their high rate of frames of 3D data, two features make them difficult as mapping sensors, namely, their restricted field of view and influences on the quality of range measurements by high dynamics in object reflectivity; in addition, currently available models suffer from poor data quality in a number of aspects. The paper first summarizes calibration and filtering approaches for improving the accuracy, precision, and robustness of ToF cameras independent of their intended usage. Then, several ego motion estimation approaches are applied or adapted, respectively, in order to provide a performance benchmark for registering ToF camera data. As a part of this, an extension to the iterative closest point algorithm has been developed that increases the robustness under restricted field of view and under larger displacements. Using an indoor environment, the paper provides results from SLAM experiments using these approaches in comparison. It turns out that the application of ToF cameras is feasible to SLAM tasks, although this type of sensor has a complex error characteristic.

Link

Lab Meeting November 22, 2010 (Alan): Temporary Maps for Robust Localization in Semi-static Environments (IROS 2010)

Title: Temporary Maps for Robust Localization in Semi-static Environments (IROS 2010)

Authors: Daniel Meyer-Delius, Jurgen Hess, Giorgio Grisetti, Wolfram Burgard

Abstract—Accurate and robust localization is essential for the successful navigation of autonomous mobile robots. The majority of existing localization approaches, however, is based on the assumption that the environment is static which does not hold for most practical application domains. In this paper, we present a localization framework that can robustly track a robot’s pose even in non-static environments. Our approach keeps track of the observations caused by unexpected objects in the environment using temporary local maps. It relies both on these temporary local maps and on a reference map of the environment for estimating the pose of the robot. Experimental results demonstrate that by exploiting the observations caused by unexpected objects our approach outperforms standard localization methods for static environments.

Link: pdf

Lab Meeting November 15( KuenHan ), 3D Reconstruction of a Moving Point from a Series of 2D Projections (ECCV 2010)

Title :3D Reconstruction of a Moving Point from a Series of 2D Projections
Author: Hyun Soo Park, Takaaki Shiratori, Iain Matthews, and Yaser Sheikh

Abstract

This paper presents a linear solution for reconstructing the 3D trajectory of a moving point from its correspondence in a collection of 2D perspective images, given the 3D spatial pose and time of capture of the cameras that produced each image. Triangulation-based solutions do not apply, as multiple views of the point may not exist at each instant in time. A geometric analysis of the problem is presented and a criterion, called reconstructibility, is defined to precisely characterize the cases when reconstruction is possible, and how accurate it can be. We apply the linear reconstruction algorithm to reconstruct the time evolving 3D structure of several real-world scenes, given a collection of non-coincidental 2D images.

Link

Lab Meeting November 15, 2010 (fish60): Unfreezing the Robot: Navigation in Dense, Interacting Crowds

Title: Unfreezing the Robot: Navigation in Dense, Interacting Crowds(IROS 2010)
Author: Peter Trautman and Andreas Krause

Abstract—In this paper, we study the safe navigation of a mobile robot through crowds of dynamic agents with uncertain trajectories. Existing algorithms suffer from the “freezing robot” problem: once the environment surpasses a certain level of complexity, the planner decides that all forward paths are unsafe, and the robot freezes in place (or performs unnecessary aneuvers) to avoid collisions. ... In this work, we demonstrate that both the individual prediction and the predictive uncertainty have little to do with the frozen robot problem. Our key insight is that dynamic agents solve the frozen robot problem by engaging in “joint collision avoidance”: They cooperatively make room to create feasible trajectories. We develop IGP, a nonparametric statistical model based on Dependent Output Gaussian Processes that can estimate crowd interaction from data. Our model naturally captures the non-Markov nature of agent trajectories, as well as their goal-driven navigation. We then show how planning in this model can be efficiently implemented using particle based inference.

Link

CMU PhD Thesis Defense: Geolocation with Range: Robustness, Efficiency and Scalability

CMU RI PhD Thesis Defense
Joseph A. Djugash
Geolocation with Range: Robustness, Efficiency and Scalability
November 05, 2010, 10:00 a.m., NSH 1507

Abstract

This thesis explores the topic of geolocation with range. A robust method for localization and SLAM (Simultaneous Localization and Mapping) is proposed. This method uses a polar parameterization of the state to achieve accurate estimates of the nonlinear and multi-modal distributions in range-only systems. Several experimental evaluations on real robots reveal the reliability of this method.

Scaling such a system to large network of nodes, increases the computational load on the system due to the increased state vector. To alleviate this problem, we propose the use of a distributed estimation algorithm based on the belief propagation framework. This method distributes the estimation task, such that each node only estimates its local network, greatly reducing the computation performed by any individual node. However, the method does not provide any guarantees on the convergence of its solution in general graphs. Convergence is only guaranteed for non-cyclic graphs (ie. trees). Thus, an extension of this approach which reduces any arbitrary graph to a spanning tree is presented. This enables the proposed decentralized localization method to provide guarantees on its convergence.

Scaling in the traditional sense involves extensions to deal with growth in the size of the operating environment. In large, feature-less environments, maintaining a globally consistent estimate of a group of mobile agents is difficult. In this thesis, a novel multi-robot coordination strategy is proposed. Based on the observability analysis of the system, the propose controller achieves the tight coordination necessary to obtain an accurate global estimate. The proposed approach is demonstrated using both simulation and experimental testing with real robots.

[LINK][PDF]

Thesis Committee
Sanjiv Singh, Chair
George Kantor
Howie Choset
Wolfram Burgard, University of Freiburg

Lab Meeting November 1, 2010 (Will): Visual Event Recognition in Videos by Learning from Web Data (CVPR 2010)

Title: Visual Event Recognition in Videos by Learning from Web Data (CVPR 2010)
Author: Lixin Duan, Dong Xu, Ivor W. Tsang, Jiebo Luo

Abstract:
We propose a visual event recognition framework for consumer domain videos by leveraging a large amount of loosely labeled web videos (e.g., from YouTube). First, we propose a new aligned space-time pyramid matching method to measure the distances between two video clips, where each video clip is divided into space-time volumes over multiple levels. We calculate the pairwise distances between any two volumes and further integrate the information from different volumes with Integer-flow Earth Mover’s Distance (EMD) to explicitly align the volumes. Second, we propose a new cross-domain learning method in order to 1) fuse the information from multiple pyramid levels and features (i.e., space-time feature and static SIFT feature) and 2) cope with the considerable variation in feature dis- tributions between videos from two domains (i.e., web do- main and consumer domain). For each pyramid level and each type of local features, we train a set of SVM classifiers based on the combined training set from two domains using multiple base kernels of different kernel types and parameters, which are fused with equal weights to obtain an average classifier. Finally, we propose a cross-domain learning method, referred to as Adaptive Multiple Kernel Learning (A-MKL), to learn an adapted classifier based on multiple base kernels and the prelearned average classifiers by minimizing both the structural risk functional and the mismatch between data distributions from two domains. Extensive experiments demonstrate the effectiveness of our proposed framework that requires only a small number of labeled consumer videos by leveraging web data.

link: [http://vc.sce.ntu.edu.sg/index_files/a-mkl.pdf]

local link: [https://pal.csie.ntu.edu.tw/pub2/Conferences/2010_CVPR/data/papers/1358.pdf#page=1]

Lab meeting Nov. 01 2010, (Chih-Chung) POMDPs for robotic tasks with mixed observability (RSS 2009)

Title:POMDPs for robotic tasks with mixed observability
Author:Sylvie C.W.Ong, Shao Wei Png, David Hsu and Wee Sun Lee.

Abstract:
Partially observable Markov decision processes
(POMDPs) provide a principled mathematical framework for
motion planning of autonomous robots in uncertain and dynamic
environments. They have been successfully applied to
various robotic tasks, but a major challenge is to scale up
POMDP algorithms for more complex robotic systems. Robotic
systems often have mixed observability: even when a robot’s
state is not fully observable, some components of the state
may still be fully observable. Exploiting this, we use a factored
model to represent separately the fully and partially observable
components of a robot’s state and derive a compact lowerdimensional
representation of its belief space. We then use this
factored representation in conjunction with a point-based algorithm
to compute approximate POMDP solutions. Separating
fully and partially observable state components using a factored
model opens up several opportunities to improve the efficiency
of point-based POMDP algorithms. Experiments show that on
standard test problems, our new algorithm is many times faster
than a leading point-based POMDP algorithm.

link

News: University of Chicago, Cornell Researchers Develop Universal Robotic Gripper

Robotic hands are usually just that -- hands -- but some researchers from the University of Chicago and Cornell University (with a little help from iRobot) have taken a decidedly different approach for their so-called universal robotic gripper. As you can see above, the gripper is actually a balloon that can conform to and grip just about any small object, and hang onto it firmly enough to pick it up. What's the secret? After much testing, the researchers found that ground coffee was the best substance to fill the balloon with -- to grab an object, the gripper simply creates a vacuum in the balloon (much like a vacuum-sealed bag of coffee), and it's then able to let go of the object just by releasing the vacuum. Simple, but it works. Head on past the break to check it out in action. [via engadget]

Lab meeting Oct. 25 2010, (David) Threat-aware Path Planning in Uncertain Urban Environments (IROS 2010)

Title: Threat-aware Path Planning in Uncertain Urban Environments

Authors: Georges S. Aoude, Brandon D. Luders, Daniel S. Levine, and Jonathan P. How

Abstract:
This paper considers the path planning problem
for an autonomous vehicle in an urban environment populated
with static obstacles and moving vehicles with uncertain intents.
We propose a novel threat assessment module, consisting of
an intention predictor and a threat assessor, which augments
the host vehicle’s path planner with a real-time threat value
representing the risks posed by the estimated intentions of
other vehicles. This new threat-aware planning approach is
applied to the CL-RRT path planning framework, used by the
MIT team in the 2007 DARPA Grand Challenge. The strengths
of this approach are demonstrated through simulation and
experiments performed in the RAVEN testbed facilities

[local copy]

[link ]

[local video]

[video]

Lab meeting Oct. 11 2010, (Shao-Chen) Consistent data association in multi-robot systems with limited communications(RSS 2010)

Title: Consistent data association in multi-robot systems with limited communications

Authors: Rosario Aragues,Eduardo Montijano, and Carlos Sagues

Abstract:

In this paper we address the data association
problem of features observed by a robot team with limited communications.
At every time instant, each robot can only exchange
data with a subset of the robots, its neighbors. Initially, each
robot solves a local data association with each of its neighbors.
After that, the robots execute the proposed algorithm to agree
on a data association between all their local observations which
is globally consistent. One inconsistency appears when chains of
local associations give rise to two features from one robot being
associated among them. The contribution of this work is the
decentralized detection and resolution of these inconsistencies.
We provide a fully decentralized solution to the problem. This
solution does not rely on any particular communication topology.
Every robot plays the same role, making the system robust to
individual failures. Information is exchanged exclusively between
neighbors. In a finite number of iterations, the algorithm finishes
with a data association which is free of inconsistent associations.
In the experiments, we show the performance of the algorithm
under two scenarios. In the first one, we apply the resolution
and detection algorithm for a set of stochastic visual maps. In
the second, we solve the feature matching between a set of images
taken by a robotic team.

[link]

Lab meeting Oct. 11th 2010, (Nicole) Improvement in Listening Capability for Humanoid Robot HRP-2(ICRA 2010)

Title: Improvement in Listening Capability for Humanoid Robot HRP-2 (ICRA2010)

Authors: Toru Takahashi, Kazuhiro Nakadai, Kazunori Komatani, Tetsuya Ogata and Hiroshi G. Okuno.

Abstract:

This paper describes improvement of sound source separation for a simultaneous automatic speech recognition (ASR) system of a humanoid robot. A recognition error in the system is caused by a separation error and interferences of other sources. In separability, an original geometric source separation (GSS) is improved. Our GSS uses a measured robot’s head related transfer function (HRTF) to estimate a separation matrix. As an original GSS uses a simulated HRTF calculated based on a distance between microphone and sound source, there is a large mismatch between the simulated and the measured transfer functions. The mismatch causes a severe degradation of recognition performance.

Faster convergence speed of separation matrix reduces separation error. Our approach gives a nearer initial separation matrix based on a measured transfer function from an optimal separation matrix than a simulated one. As a result, we expect that our GSS improves the convergence speed. Our GSS is also able to handle an adaptive step-size parameter.

These new features are added into open source robot audition software (OSS) called”HARK” which is newly updated as version 1.0.0. The HARK has been installed on a HRP-2 humanoid with an 8-element microphone array. The listening capability of HRP-2 is evaluated by recognizing a target speech signal which is separated from a simultaneous speech signal by three talkers. The word correct rate (WCR) of ASR improves by 5 points under normal acoustic environments and by 10 points under noisy environments. Experimental results show that HARK 1.0.0 improves the robustness against noises.

[link]

Lab meeting Oct. 11th 2010, (Andi) Dynamic 3D Scene Analysis for Acquiring Articulated Scene Models

Dynamic 3D Scene Analysis for Acquiring Articulated Scene Models

ICRA 2010

Agnes Swadzba, Niklas Beuter, Sven Wachsmuth, and Franz Kummert

Abstract:

In this paper we present a new system for a mobile robot to generate an articulated scene model by analyzing complex dynamic 3D scenes. The system extracts essential knowledge about the foreground, like moving persons, and the background, which consists of all visible static scene parts. In contrast to other 3D reconstruction approaches, we suggest to additionally distinguish between static parts, like walls, and movable objects like chairs or doors. The discrimination supports the reconstruction process and additionally, delivers important information about interaction objects. Here, the movable object detection is realized object independent by analyzing changes in the scenery. Furthermore, in the proposed system the background scene is feedbacked to the tracking part yielding a much better tracking and detection result which improves again the 3D reconstruction. We show in our experiments that we are able to provide a sound background model and to extract simultaneously persons and object regions representing chairs, doors, and even smaller movable objects.

paper link

News: Google's Self-Driving Cars

Google: We've Been Secretly Building And Testing Robot Cars That Drive Themselves
By Sebastian Thrun, Google
Read the full article.

Google Cars Drive Themselves, in Traffic
By John Markoff, The New York Times 
Read the full article.

News: MIT Media Lab Medical Mirror

MIT Medical Media Lab Mirror tells your pulse with a webcam

Mirror mirror on the wall, who has the highest arterial palpation of them all? If you went to MIT you might be able to answer that question thanks to the work of grad student Ming-Zher Poh, who has found a way to tell your pulse with just a simple webcam and some software. By looking at minute changes in the brightness of the face, the system can find the beating of your heart even at a low resolution, comparable to the results of a traditional FDA-approved pulse monitor. Right now the mirror above is just a proof of concept, but the idea is that the hospital beds or surgery rooms of tomorrow might be able to monitor a patient's pulse without requiring any wires or physical contact, encouraging news for anyone who has ever tried to sleep whilst wearing a heart monitor. [via Engadget]

Lab Meeting October 4th, 2010 (Jeff): Progress Report

I will represent the progress in RFID SLAM with some indexing value to show the performance.

Lab Meeting October 4th, 2010(KuoHuei): progress report

I will present the Neighboring Object Interacting Tracking, including modeling, learning, inference, and some results.

Lab Meeting September 27, 2010 (Wang Li): Monocular 3D Pose Estimation and Tracking by Detection (CVPR 2010)

Monocular 3D Pose Estimation and Tracking by Detection

Mykhaylo Andriluka
Stefan Roth
Bernt Schiele

Abstract
Automatic recovery of 3D human pose from monocular
image sequences is a challenging and important research
topic with numerous applications. Although current methods
are able to recover 3D pose for a single person in controlled
environments, they are severely challenged by realworld
scenarios, such as crowded street scenes. To address
this problem, we propose a three-stage process building on
a number of recent advances. The first stage obtains an initial
estimate of the 2D articulation and viewpoint of the person
from single frames. The second stage allows early data
association across frames based on tracking-by-detection. The third and
final stage uses those tracklet-based estimates as robust image
observations to reliably recover 3D pose. We demonstrate
state-of-the-art performance on the HumanEva II
benchmark, and also show the applicability of our approach
to articulated 3D tracking in realistic street conditions.

Paper Link

Lab Meeting September 20, 2010 (Kuen-Han): Scale Drift-Aware Large Scale Monocular SLAM (RSS 2010)

Title: Scale Drift-Aware Large Scale Monocular SLAM

Author: Hauke Strasdat, J.M.M. Montiel, Andrew J. Davison

Abstract—State of the art visual SLAM systems have recently
been presented which are capable of accurate, large-scale and
real-time performance, but most of these require stereo vision.
Important application areas in robotics and beyond open up
if similar performance can be demonstrated using monocular
vision, since a single camera will always be cheaper, more
compact and easier to calibrate than a multi-camera rig.
With high quality estimation, a single camera moving through
a static scene of course effectively provides its own stereo
geometry via frames distributed over time. However, a classic
issue with monocular visual SLAM is that due to the purely
projective nature of a single camera, motion estimates and map
structure can only be recovered up to scale. Without the known
inter-camera distance of a stereo rig to serve as an anchor, the
scale of locally constructed map portions and the corresponding
motion estimates is therefore liable to drift over time.
In this paper we describe a new near real-time visual SLAM
system which adopts the continuous keyframe optimisation approach
of the best current stereo systems, but accounts for
the additional challenges presented by monocular input. In
particular, we present a new pose-graph optimisation technique
which allows for the efficient correction of rotation, translation
and scale drift at loop closures. Especially, we describe the
Lie group of similarity transformations and its relation to the
corresponding Lie algebra. We also present in detail the system’s
new image processing front-end which is able accurately to track
hundreds of features per frame, and a filter-based approach
for feature initialisation within keyframe-based SLAM. Our
approach is proven via large-scale simulation and real-world
experiments where a camera completes large looped trajectories.

link