Detail publikace
EnMS: Early non-Maxima Suppression
HEROUT, A. HRADIŠ, M. ZEMČÍK, P.
Originální název
EnMS: Early non-Maxima Suppression
Anglický název
EnMS: Early non-Maxima Suppression
Jazyk
en
Originální abstrakt
Detection of objects in images using statistical classifiers is a well studied and documented technique. Different applications of such detectors often require selection of the image position with the highest response of the detector -- they perform non-maxima suppression. This article introduces the concept of Early non-Maxima Suppression, which aims to reduce necessary computations by making the non-Maxima Suppression decision early based on incomplete information provided by a partially evaluated classifier. We show that the error of one such speculative decision with respect to a decision made based on response of the complete classifier can be estimated by collecting statistics on unlabeled data. The article then considers a sequential strategy of multiple early non-Maxima suppression tests which follows the structure of soft-cascade detectors commonly used for object detection. We also show that an optimal (fastest for requested error rate) suppression strategy can be created by a novel variant of Wald's sequential probability ratio test (SPRT) which we call the Conditioned SPRT, CSPRT. Experimental results show that the Early non-Maxima Suppression significantly reduces amount of computation in the case of object localization while the error rates are limited to low predefined values. The proposed approach notably outperforms the state-of-the-art detectors based on WaldBoost. The potential applications of the early non-Maxima suppression approach are not limited to object localization and could be applied wherever the goal is to find the strongest response of a classifier among a set of classified samples.
Anglický abstrakt
Detection of objects in images using statistical classifiers is a well studied and documented technique. Different applications of such detectors often require selection of the image position with the highest response of the detector -- they perform non-maxima suppression. This article introduces the concept of Early non-Maxima Suppression, which aims to reduce necessary computations by making the non-Maxima Suppression decision early based on incomplete information provided by a partially evaluated classifier. We show that the error of one such speculative decision with respect to a decision made based on response of the complete classifier can be estimated by collecting statistics on unlabeled data. The article then considers a sequential strategy of multiple early non-Maxima suppression tests which follows the structure of soft-cascade detectors commonly used for object detection. We also show that an optimal (fastest for requested error rate) suppression strategy can be created by a novel variant of Wald's sequential probability ratio test (SPRT) which we call the Conditioned SPRT, CSPRT. Experimental results show that the Early non-Maxima Suppression significantly reduces amount of computation in the case of object localization while the error rates are limited to low predefined values. The proposed approach notably outperforms the state-of-the-art detectors based on WaldBoost. The potential applications of the early non-Maxima suppression approach are not limited to object localization and could be applied wherever the goal is to find the strongest response of a classifier among a set of classified samples.
Dokumenty
BibTex
@article{BUT76262,
author="Adam {Herout} and Michal {Hradiš} and Pavel {Zemčík}",
title="EnMS: Early non-Maxima Suppression",
annote="Detection of objects in images using statistical classifiers is a well studied
and documented technique. Different applications of such detectors often require
selection of the image position with the highest response of the detector -- they
perform non-maxima suppression. This article introduces the concept of Early
non-Maxima Suppression, which aims to reduce necessary computations by making the
non-Maxima Suppression decision early based on incomplete information provided by
a partially evaluated classifier. We show that the error of one such speculative
decision with respect to a decision made based on response of the complete
classifier can be estimated by collecting statistics on unlabeled data. The
article then considers a sequential strategy of multiple early non-Maxima
suppression tests which follows the structure of soft-cascade detectors commonly
used for object detection. We also show that an optimal (fastest for requested
error rate) suppression strategy can be created by a novel variant of Wald's
sequential probability ratio test (SPRT) which we call the Conditioned SPRT,
CSPRT. Experimental results show that the Early non-Maxima Suppression
significantly reduces amount of computation in the case of object localization
while the error rates are limited to low predefined values. The proposed approach
notably outperforms the state-of-the-art detectors based on WaldBoost. The
potential applications of the early non-Maxima suppression approach are not
limited to object localization and could be applied wherever the goal is to find
the strongest response of a classifier among a set of classified samples.",
address="NEUVEDEN",
chapter="76262",
edition="NEUVEDEN",
howpublished="print",
institution="NEUVEDEN",
number="2",
volume="2012",
year="2012",
month="may",
pages="121--132",
publisher="NEUVEDEN",
type="journal article - other"
}