A Caltech Library Service

Fast Terrain Classification Using Variable-Length Representation for Autonomous Navigation

Angelova, Anelia and Matthies, Larry and Helmick, Daniel and Perona, Pietro (2007) Fast Terrain Classification Using Variable-Length Representation for Autonomous Navigation. In: IEEE Conference on Computer Vision and Pattern Recognition, 2007. CVPR '07. IEEE , Piscataway, NJ. ISBN 1-4244-1179-3.

[img] PDF - Accepted Version
See Usage Policy.

[img] PDF - Published Version
See Usage Policy.


Use this Persistent URL to link to this item:


We propose a method for learning using a set of feature representations which retrieve different amounts of information at different costs. The goal is to create a more efficient terrain classification algorithm which can be used in real-time, onboard an autonomous vehicle. Instead of building a monolithic classifier with uniformly complex representation for each class, the main idea here is to actively consider the labels or misclassification cost while constructing the classifier. For example, some terrain classes might be easily separable from the rest, so very simple representation will be sufficient to learn and detect these classes. This is taken advantage of during learning, so the algorithm automatically builds a variable-length visual representation which varies according to the complexity of the classification task. This enables fast recognition of different terrain types during testing. We also show how to select a set of feature representations so that the desired terrain classification task is accomplished with high accuracy and is at the same time efficient. The proposed approach achieves a good trade-off between recognition performance and speedup on data collected by an autonomous robot.

Item Type:Book Section
Related URLs:
URLURL TypeDescription
Perona, Pietro0000-0002-7583-5809
Additional Information:© 2007 IEEE. This research was carried out by the Jet Propulsion Laboratory, California Institute of Technology with funding from the NASA's Mars Technology Program. We thank Max Bajracharya and the anonymous reviewers for providing very useful comments on the paper.
Funding AgencyGrant Number
Record Number:CaltechAUTHORS:20150903-130801389
Persistent URL:
Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:60053
Deposited By: Caroline Murphy
Deposited On:16 Sep 2015 21:14
Last Modified:10 Nov 2021 22:29

Repository Staff Only: item control page