Workshop on

Visual Localization for Mobile Platforms

28 June 2008

Anchorage, Alaska

To be held in conjunction with CVPR 2008

News!!!

Mar 19, Final deadline for papers Mar 24, 2008 11:30 PM PST

Mar 13, 2008: Paper submission deadline now extended to 17 March 11:30 PM PST

Feb 29, 2008 : Paper submissions now open

May 23, 2008 Final Schedule Published

Motivation

Recent advances have shown the possibility of doing real time vision based localization over extended regions, indoors and outdoors. There has been an explosion of research in visual localization driven from both the fields of robot mapping and vision research for mobile robots. This workshop will be useful in bringing together researchers to talk about
the successes and limitations of their techniques, and how they will approach the significant problems still remaining.

A particular motivation is to see how different systems perform on standard datasets, since it is often difficult to judge comparative performance from separate papers. We will provide several types of datasets, both indoor and outdoor to the researchers. We will ask
presenters to exercise their systems on the datasets, and fill out a performance worksheet, which will be analyzed and distributed at the workshop.

Topics

Topics relevant for this workshop include but are not limited to

Visual Odometry
Visual SLAM
Monocular and multi camera localization
Mapping using omni directional cameras
Features for visual tracking
Wide baseline scene matching
Landmarks recognition
Visual loop closure in small and large environments
Integration of vision sensors with exogenous sensors (IMUs, motion constraints,
etc.)
Stability and robustness of SLAM in small environments
Large-scale and outdoor mapping

Final Program

June 28, 2008

8:00 – 8:15 Welcome Remarks

Oral Session1

8:15 – 9:00 Invited Speaker Andrew Comport Accurate and Robust 3D Visual odometry

9:00 – 9:30 W14.01 Dense Linear-Time Correspondences for Tracking

9:30 – 10:00 W14.02 A Dual-Layer Estimator Architecture for Long-term Localization

10:00-10:30 Coffee Break

10:30-12:00 Poster Session

W14.04 Improving the Selection and Detection of Visual Landmarks Through Object Tracking

W14.05 3D Tracking Using On-Line Keypoint Learning for Mobile Augmented Reality

W14.06 Visual Map Matching and Localization Using a Global Feature Map

W14.07 Autonomous Navigation and Mapping using Monocular Low-Resolution Grayscale Vision

W14.08 Spatio-Temporal Consistency and Distributivity as Qualities of Features

W14.09 Improving RANSAC for Fast Landmark Recognition

W14.10 Incremental estimation without specifying a-priori covariance matrices for the novel parameters

12:00 -1:30 Lunch Break

Oral Session2

1:30 – 2:45 Guest Speaker Motilal Agrawal -- Features for Real Time Visual Odometry

2:45 – 3:15 W14.03 Camera Localization and Building Reconstruction from Single Monocular Images

3:15- 3:45 Coffee Break

Oral Session3

3:45 - 4:15 W14.11 Visual Loop Closing Using Extremely Randomized Trees

4:15 – 4:45 W14.12 Boosting Descriptors Condensed from Video Sequences for Place Recognition

4:45 – 5:15 W14.13 Visual Detection of Lintel-Occluded Doors from a Single Image