CaltechAUTHORS
  A Caltech Library Service

Fast Feature Pyramids for Object Detection

Dollár, Piotr and Appel, Ron and Belongie, Serge and Perona, Pietro (2014) Fast Feature Pyramids for Object Detection. IEEE Transactions on Pattern Analysis and Machine Intelligence, 36 (8). pp. 1532-1545. ISSN 0162-8828. http://resolver.caltech.edu/CaltechAUTHORS:20140904-111257556

[img]
Preview
PDF - Submitted Version
See Usage Policy.

3707Kb

Use this Persistent URL to link to this item: http://resolver.caltech.edu/CaltechAUTHORS:20140904-111257556

Abstract

Multi-resolution image features may be approximated via extrapolation from nearby scales, rather than being computed explicitly. This fundamental insight allows us to design object detection algorithms that are as accurate, and considerably faster, than the state-of-the-art. The computational bottleneck of many modern detectors is the computation of features at every scale of a finely-sampled image pyramid. Our key insight is that one may compute finely sampled feature pyramids at a fraction of the cost, without sacrificing performance: for a broad family of features we find that features computed at octave-spaced scale intervals are sufficient to approximate features on a finely-sampled pyramid. Extrapolation is inexpensive as compared to direct feature computation. As a result, our approximation yields considerable speedups with negligible loss in detection accuracy. We modify three diverse visual recognition systems to use fast feature pyramids and show results on both pedestrian detection (measured on the Caltech, INRIA, TUD-Brussels and ETH data sets) and general object detection (measured on the PASCAL VOC). The approach is general and is widely applicable to vision algorithms requiring fine-grained multi-scale analysis. Our approximation is valid for images with broad spectra (most natural images) and fails for images with narrow band-pass spectra (e.g., periodic textures).


Item Type:Article
Related URLs:
URLURL TypeDescription
http://dx.doi.org/10.1109/TPAMI.2014.2300479 DOIArticle
http://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=6714453PublisherArticle
ORCID:
AuthorORCID
Perona, Pietro0000-0002-7583-5809
Additional Information:© 2014 IEEE. Manuscript received 14 Feb. 2013; revised 15 Dec. 2013; accepted 8 Jan. 2014. Date of publication 15 Jan. 2014; date of current version 10 July 2014. The authors thank Peter Welinder and Rodrigo Benenson for helpful comments and suggestions. Piotr Dollár, Ron Appel, and Pietro Perona were supported by MURI-ONR N00014-10-1-0933 and ARO/JPL-NASA Stennis NAS7.03001. Ron Appel was also supported by NSERC 420456-2012 and The Moore Foundation. Serge Belongie was supported by the US National Science Foundation (NSF) CAREER Grant 0448615, MURI-ONR N00014-08-1- 0638 and a Google Research Award.
Funders:
Funding AgencyGrant Number
MURI-ONRN00014-10-1-0933
ARO/JPL-NASA StennisNAS7.03001
NSERC420456-2012
Gordon and Betty Moore FoundationUNSPECIFIED
NSF0448615
Office of Naval Research (ONR)N00014-08-1-0638
Google Research AwardUNSPECIFIED
Subject Keywords:Visual features; object detection; image pyramids; pedestrian detection; natural image statistics; real-time systems
Record Number:CaltechAUTHORS:20140904-111257556
Persistent URL:http://resolver.caltech.edu/CaltechAUTHORS:20140904-111257556
Official Citation:Dollar, P.; Appel, R.; Belongie, S.; Perona, P., "Fast Feature Pyramids for Object Detection," Pattern Analysis and Machine Intelligence, IEEE Transactions on , vol.36, no.8, pp.1532,1545, Aug. 2014 doi: 10.1109/TPAMI.2014.2300479 URL: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6714453&isnumber=6848879
Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:49239
Collection:CaltechAUTHORS
Deposited By: Tony Diaz
Deposited On:04 Sep 2014 18:54
Last Modified:08 May 2017 22:41

Repository Staff Only: item control page