A multi-expert approach for robust face detection
A multi-expert approach for robust face detection
复制标题
DOI：
10.1016/j.patcog.2005.11.020
复制
发表时间：
2004-09
期刊：
影响因子：
--
通讯作者：
Linlin Huang;A. Shimizu
中科院分区：
文献类型：
--
作者：
Linlin Huang;A. Shimizu
检测精度和检测速度都是为实际应用开发健壮的人脸检测系统的主要关注点。为此，我们提出了一种多专家级联和并行相结合的人脸检测方法。我们设计了三种检测专家，分别采用不同的局部图像特征表示方法：二维Haar小波、梯度方向和Gabor滤波。这三种特征使用相同的分类模型进行分类，即在约简特征子空间上使用多项式神经网络(PNN)。为了提高检测速度，将检测专家分成多个阶段，在进行阶段使用简单的专家，在后续阶段使用复杂的专家。同时，将每个专家的输出与其先前专家的输出相结合，以提高检测的准确率。通过对大量图像的实验，证明了多专家方法的有效性。所获得的检测结果优于最好的个体专家和最先进的方法，同时检测速度快。
Both detection accuracy and speed are of major concerns in developing a robust face detection system for real-world applications. To this end, we propose a robust face detection approach by combining multiple experts in both cascade and parallel manner. We design three detection experts which employ different feature representation schemes of local images: 2D Haar wavelet, gradient direction, and Gabor filter. The three features are classified using the same classification model, namely, a polynomial neural network (PNN) on reduced feature subspace. The detection experts are used in multiple stages with simple ones in proceeding stages and complex ones in succeeding stages for improving detection speed. Meanwhile, the output of each expert is combined with the outputs of its preceding experts to improve detection accuracy. The effectiveness of the multi-expert approach has been demonstrated in experiments on a large number of images. The obtained detection results are superior to the best individual expert and state-of-the-art approaches while the detection speed is fast.