Matrix-Structural Learning (MSL) of Cascaded Classifier from Enormous Training Set

Yan, Shengye; Shan, Shiguang; Chen, Xilin; Gao, Wen; Chen, Jie

doi:10.1109/cvpr.2007.383156

Cited by 22 publications

(10 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We use Matrix-Structural Learning (MSL) [28] to learn from the training sets, which is a cascade learning method to deal with enormous training set. In cascade learning, Minimum detection rate and maximum false alarm rate of feature-centric and window-centric cascade are both set to 0.9999 and 0.4 respectively.…”

Section: Methodsmentioning

confidence: 99%

“…Obviously, our method gets the best results. Bourdev [25] Schneiderman [23] Viola [16] Huang [21] Yan [28] Our method Figure 13: ROC curves on CMU+MIT frontal face set.…”

Section: Experiments On Frontal Face Detectionmentioning

confidence: 99%

“…After this seminal work, many improved versions were proposed. Mostly of them focused on alternatives to AdaBoost [12, 13, 14, 15, 16, 17, 18 and 19], Haar features [19,20], coarse-to-fine architecture [21,22,23,24] and the optimization tuning of the cascade architecture [17,18,26,27,28].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Locally Assembled Binary (LAB) feature with feature-centric cascade for fast and accurate face detection

Yan¹,

Shan²,

Chen³

et al. 2008

2008 IEEE Conference on Computer Vision and Pattern Recognition

Self Cite

View full text Add to dashboard Cite

show abstract

Section: Methodsmentioning

confidence: 99%

“…Obviously, our method gets the best results. Bourdev [25] Schneiderman [23] Viola [16] Huang [21] Yan [28] Our method Figure 13: ROC curves on CMU+MIT frontal face set.…”

Section: Experiments On Frontal Face Detectionmentioning

confidence: 99%

See 1 more Smart Citation

Locally Assembled Binary (LAB) feature with feature-centric cascade for fast and accurate face detection

Yan¹,

Shan²,

Chen³

et al. 2008

2008 IEEE Conference on Computer Vision and Pattern Recognition

Self Cite

View full text Add to dashboard Cite

show abstract

“…For all the images, the face detection method [41] is applied to locate the face region from the input images, and then, all the face regions are normalized to the same size of 32 × 32. Finally, histogram equalization is used to reduce the influence of lighting variations.…”

Section: Methodsmentioning

confidence: 99%

Head Yaw Estimation From Asymmetry of Facial Appearance

Shan

Chen

et al. 2008

IEEE Trans. Syst., Man, Cybern. B

View full text Add to dashboard Cite

Abstract-This paper proposes a novel method to estimate the head yaw rotations based on the asymmetry of 2-D facial appearance. In traditional appearance-based pose estimation methods, features are typically extracted holistically by subspace analysis such as principal component analysis, linear discriminant analysis (LDA), etc., which are not designed to directly model the pose variations. In this paper, we argue and reveal that the asymmetry in the intensities of each row of the face image is closely relevant to the yaw rotation of the head and, at the same time, evidently insensitive to the identity of the input face. Specifically, to extract the asymmetry information, 1-D Gabor filters and Fourier transform are exploited. LDA is further applied to the asymmetry features to enhance the discrimination ability. By using the simple nearest centroid classifier, experimental results on two multipose databases show that the proposed features outperform other features. In particular, the generalization of the proposed asymmetry features is verified by the impressive performance when the training and the testing data sets are heterogeneous.Index Terms-Fourier transform, Gabor filters, head yaw estimation, linear discriminant analysis (LDA), nearest centroid (NC) classifier.

show abstract

“…So, the often applied LUT classifiers only have moderate number of partitions, such as less than 13 [9], 16 [19], 32 [4], [5], [11], etc. To further improve the generalization performance of LUT based classifiers, much more training samples are needed to train a high performance LUT classifiers based object detector [3], [4] as compared to decision stump based one [8]. How to design efficient LUT classifiers for fast and robust object detection?…”

mentioning

confidence: 99%

Smoothing LUT classifiers for robust face detection

Wen

Xiong

2013

2013 IEEE Third International Conference on Information Science and Technology (ICIST)

View full text Add to dashboard Cite

Look-up table (LUT) classifiers are often used to construct concise classifiers for rapid object detection due to their favorable convergent ability. However, their poor generalization ability imposes restrictions on their applications. A novel improvement to LUT classifiers is proposed in this paper where the new confident of each partition is recalculated by smoothing the old ones within its neighbor partitions. The new confidents are more generalizable than the old ones because each of the new predicts is supported by more training samples and the high frequency components in the old predict sequences are suppressed greatly through smoothing operation. Both weight sum smoothing method and confident smoothing method are introduced here, which all bring negligible extra computation cost in training time and no extra cost in test time. Experimental results in the domain of upright frontal face detection using smoothed LUT classifiers with identical smoothing width and smoothing factor for all partitions based on Haar-like rectangle features show that smoothed LUT classifiers generalize much better and also converge more or less worse than unsmooth LUT classifiers. Specifically, smoothed LUT classifiers can delicately balance between generalization ability and convergent ability through carefully set smoothing parameters. I. INTRODUCTIONRIGINATED from domain-partitioning confidencerated hypotheses [1], Look-up table (LUT) classifiers [5], [2], [17] (response binning classifiers) are often used to construct very concise classifiers through Real AdaBoost [1] for rapid object detection and classification due to their favorable convergent ability, such as face detection [2]-[4], [9]-[16], [19]-[21], gender classification [5], pedestrian detection [6], [22], [23], incident detection [7], etc. The LUT classifiers used in these papers are somewhat revised by the authors to meet their needs, such as using adaptive (much less) partitions for resource-constrained devices [9], discarding negative responses and adding global predicts [7], online adjusting the confidents for online learning [10], using 0-1 version LUT classifiers to meet the needs of Discrete AdaBoost other than Real AdaBoost [5], applying Bayesian error based LUT classifiers for higher performance [4], etc. Besides, various criterions are used to choose optimal LUT classifiers to form strong classifiers, such as minimizing Bhattacharyya distance [1], minimizing Kullback-Leibler divergence [19], minimizing symmetric Jensen-Shannon divergence [12], etc. However, the poor generalization ability of LUT classifiers Manuscript received December 31, 2012 J Wen, corresponding author, is with the College

show abstract

Matrix-Structural Learning (MSL) of Cascaded Classifier from Enormous Training Set

Cited by 22 publications

References 21 publications

Locally Assembled Binary (LAB) feature with feature-centric cascade for fast and accurate face detection

Locally Assembled Binary (LAB) feature with feature-centric cascade for fast and accurate face detection

Head Yaw Estimation From Asymmetry of Facial Appearance

Smoothing LUT classifiers for robust face detection

Contact Info

Product

Resources

About