Empirical Analysis of Detection Cascades of Boosted Classifiers for Rapid Object Detection

By Rainer Lienhart, Alexander Kuranov, Vadim Pisarevsky. DAGM-Symposium, 2003

971 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/67c63f6b3eeb0b619d6306fcf1c4b144710f79b0

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Vision and Pattern Recognition

Citations in the corpus

A survey on face detection in the wild: Past, present and future (2015, methodology): “RepresentativeWorks General AdaBoost[17] boosting RealBoost[59] [60] [61] [62] schemes GentleBoost[107] [66] FloatBoost[59] Reuseprevious Boostingchain[110] nodes'results Nestedcascade[61] Introduce AsymmetricBoosting[111] [112] [113] asymmetry Linearasymmetricclassifier[116] SetintermediateFixednod”
A Survey of Recent Advances in Face Detection (2010, methodology): “General AdaBoost [92] boosting RealBoost [46, 6, 101, 62] schemes GentleBoost [48, 8] FloatBoost [46]”
Sharing features: efficient boosting procedures for multiclass object detection (2004, methodology): “We chose to base our algorithm on the version of boosting called " gentleboost " [7], because it is simple to implement , numerically robust, and has been shown experimentally [11] to outperform other boosting variants for the face detection task.”
An Efficient Shared Multi-Class Detection Cascade (2008, methodology): “Instead of weak classifiers built on simple thresholding of the feature value, we use a probabilistic classification model (see Fig.”
Fully automatic facial feature point detection using Gabor feature based boosted classifiers (2005, positive): “It is simple to implement, it is numerically robust and it has been shown experimentally to outperform (with respect to detection accuracy) other boosting variants for the face detection task [21].”
Privacy Visor: Method Based on Light Absorbing and Reflecting Properties for Preventing Face Image Detection (2013, methodology): “To determine the composition of Haarlike features trained by supervised learning, we used an OpenCV [12] example cascade that had been trained in advance by using 5000 positive and 3000 negative images [13].”
Towards semantic maps for mobile robots (2008, background): “According to [23], it is the most successful learning procedure tested for face detection applications.”
Boosted local structured HOG-LBP for object localization (2011, methodology): “In our approach, we choose GentleBoost [11, 22] mainly because GentleBoost uses Newton stepping rather than exact optimization at each step, thus it outperforms other boosting methods especially when the training data is noisy and has outliers [13].”
A Survey on Object Detection in Optical Remote Sensing Images (2016, neutral): “Consequently, we here focus on a brief review of the Gentle AdaBoost algorithm.”