Joint Cascade Optimization Using A Product Of Boosted Classifiers
نویسندگان
چکیده
The standard strategy for efficient object detection consists of building a cascade composed of several binary classifiers. The detection process takes the form of a lazy evaluation of the conjunction of the responses of these classifiers, and concentrates the computation on difficult parts of the image which cannot be trivially rejected. We introduce a novel algorithm to construct jointly the classifiers of such a cascade, which interprets the response of a classifier as the probability of a positive prediction, and the overall response of the cascade as the probability that all the predictions are positive. From this noisy-AND model, we derive a consistent loss and a Boosting procedure to optimize that global probability on the training set. Such a joint learning allows the individual predictors to focus on a more restricted modeling problem, and improves the performance compared to a standard cascade. We demonstrate the efficiency of this approach on face and pedestrian detection with standard data-sets and comparisons with reference baselines.
منابع مشابه
Improving 2D Boosted Classifiers Using Depth LDA Classifier for Robust Face Detection
Face detection plays an important role in Human Robot Interaction. Many of services provided by robots depend on face detection. This paper presents a novel face detection algorithm which uses depth data to improve the efficiency of a boosted classifier on 2D data for reduction of false positive alarms. The proposed method uses two levels of cascade classifiers. The classifiers of the first lev...
متن کاملCascade AdaBoost Classifiers with Stage Features Optimization for Cellular Phone Embedded Face Detection System
In this paper, we propose a novel feature optimization method to build a cascade Adaboost face detector for real-time applications on cellular phone, such as teleconferencing, user interfaces, and security access control. AdaBoost algorithm selects a set of features and combines them into a final strong classifier. However, conventional AdaBoost is a sequential forward search procedure using th...
متن کاملDOOMRED: A New Optimization Technique for Boosted Cascade Detectors on Enforced Training Set
We propose a new method to optimize the completely-trained boosted cascade detector on an enforced training set. Recently, due to the accuracy and real-time characteristics of boosted cascade detectors like the Adaboost, a lot of variant algorithms have been proposed to enhance the performance given a fixed number of training data. And, most of algorithms assume that a given training set well e...
متن کاملMulticlass Adaboost and Coupled Classifiers for Object Detection
Building robust and fast multiclass object detection systems is a important goal of computer vision. In the present paper we extend the well-known work of Viola and Jones on boosted cascade classifiers to the multiclass case with the goal of building multiclass and multiview object detectors. We propose to use nested cascades of multiclass boosted classifiers and we introduce the concept of cou...
متن کامل