Feature reduction for efficient object detection via l1-norm latent SVM

Authors:
Min Tan;Yueming Wang;Gang Pan
Affiliations:
College of Computer Science and Technology, Zhejiang Univ., China;Qiushi Academy for Advanced Studies, Zhejiang Univ., China;College of Computer Science and Technology, Zhejiang Univ., China
Venue:
IScIDE'12 Proceedings of the third Sino-foreign-interchange conference on Intelligent Science and Intelligent Data Engineering
Year:
2012

Citing 7
Cited 0

Histograms of Oriented Gradients for Human Detection

CVPR '05 Proceedings of the 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05) - Volume 1 - Volume 01
Automatic 3D face verification from range data

ICME '03 Proceedings of the 2003 International Conference on Multimedia and Expo - Volume 3 (ICME '03) - Volume 03
3D Face Recognition by Local Shape Difference Boosting

ECCV '08 Proceedings of the 10th European Conference on Computer Vision: Part I
Sparse Online Learning via Truncated Gradient

The Journal of Machine Learning Research
Survey of Pedestrian Detection for Advanced Driver Assistance Systems

IEEE Transactions on Pattern Analysis and Machine Intelligence
Object Detection with Discriminatively Trained Part-Based Models

IEEE Transactions on Pattern Analysis and Machine Intelligence
A Comparison of Optimization Methods and Software for Large-scale L1-regularized Linear Classification

The Journal of Machine Learning Research

Quantified Score

Hi-index	0.00

Visualization

Abstract

The deformable part model is one of the most effective methods for object detection. However, it simultaneously computes the scores for a holistic filter and several part filters in a relatively high-dimensional feature space, which leads to low computational efficiency. This paper proposes an approach to select compact and effective features by learning a sparse deformable part model using L1-norm latent SVM. A stochastic truncated sub-gradient descent method is presented to solve the L1-norm latent SVM problem. Extensive experiments are conducted on the INRIA and PASCAL VOC 2007 datasets. Compared with the feature used in L2-norm latent SVM, a highly compact feature in our method can reach the state-of-the-art performance. The feature dimensionality is reduced to 12% in the INRIA dataset and less than 30% in most PASCAL VOC 2007 datasets. At the same time, the average precisions (AP) have almost no drop using the reduced feature. With our method, the speed of the detection score computation is faster than that of the L2-norm latent SVM method by 3 times. When the cascade strategy is applied, it can be further speeded up by about an order of magnitude.