AI帮你理解科学

AI 生成解读视频

AI抽取解析论文重点内容自动生成视频


pub
生成解读视频

AI 溯源

AI解析本论文相关学术脉络


Master Reading Tree
生成 溯源树

AI 精读

AI抽取本论文的概要总结


微博一下
In the limit with a single training image per subject, Extended Sparse Representation-Based Classifier still works effectively and generalizes well to large-scale databases using the intraclass variant dictionary constructed from generic subjects that are not in the gallery set

Extended SRC: undersampled face recognition via intraclass variant dictionary.

IEEE Trans. Pattern Anal. Mach. Intell., no. 9 (2012): 1864-1870

引用526|浏览31
WOS SCOPUS EI

摘要

Sparse Representation-Based Classification (SRC) is a face recognition breakthrough in recent years which has successfully addressed the recognition problem with sufficient training images of each gallery subject. In this paper, we extend SRC to applications where there are very few, or even a single, training images per subject. Assuming...更多

代码

数据

0
简介
  • WITHIN the last two decades, face recognition systems were known to be critically dependent on discriminative feature extraction methods, such as Fisherfaces [1], [2], [3] and Laplacianfaces [4], [5], [6].
  • Wright et al have demonstrated that, once the test image can be approximated by a sparse linear combination of the training images, the choice of feature space is no longer critical [7]
  • This surprising claim is supported by the experimental results that Sparse Representation-Based Classification (SRC) with random projections-based features can outperform a number of conventional face recognition schemes, such as the nearest-neighbor classifier with Fisherfaces and Laplacianfaces-based features.
  • This is often called the undersampled problem of face recognition, which has become one of the challenges in realworld applications
重点内容
  • WITHIN the last two decades, face recognition systems were known to be critically dependent on discriminative feature extraction methods, such as Fisherfaces [1], [2], [3] and Laplacianfaces [4], [5], [6]
  • Experimental results on the AR [9] and FERET [10] databases show that the usage of intraclass variant dictionary can largely improve the sparse representation-based face recognition accuracy
  • We present experiments on publicly available databases for face recognition to demonstrate the efficacy of the proposed Extended Sparse Representation-Based Classifier (ESRC)
  • Both Sparse Representation-Based Classification (SRC) and ESRC use the Homotopy2 method [15], [11] to solve the ‘1-minimization problem with the error tolerance " 1⁄4 0:05 and identical parameters3 so that the performance difference will be solely induced by the adoption of intraclass variant dictionary
  • The superiority of ESRC appears to be more significant as the number of training images decreases
  • In the limit with a single training image per subject, ESRC still works effectively and generalizes well to large-scale databases using the intraclass variant dictionary constructed from generic subjects that are not in the gallery set
结果
  • Experimental results on the

    AR [9] and FERET [10] databases show that the usage of intraclass variant dictionary can largely improve the sparse representation-based face recognition accuracy.
  • Gabor feature-based classification using ESRC yields 99 percent accuracy on the AR data set, and LBP feature-based classification using ESRC achieves a 92.3 percent recognition rate on the most challenging FERET dup2 probe set
  • These excellent results suggest that once the dictionary is properly constructed, SRC algorithms can generalize well to the large-scale face recognition problem, even when there is only a single training image per subject.
  • The authors selected a dimension of 540 for Pixel and Gabor-based randomfaces, and a resolution of 27 Â 20 for downsampled images
结论
  • The experiments suggest a number of conclusions: 1.
  • When the training images of each class are insufficient to linearly represent the testing variability, ESRC raises the recognition rates of SRC by using the intraclass variant dictionary.
  • 2. In the limit with a single training image per subject, ESRC still works effectively and generalizes well to large-scale databases using the intraclass variant dictionary constructed from generic subjects that are not in the gallery set.
  • For both SRC and ESRC, local features such as Gaborbased features and LBP features yield much better recognition rates than pixel-based features, which suggests that their invariant properties make the samples of each class more constrained to a linear subspace
表格
  • Table1: Comparative Error Rates of SRC and ESRC on the AR Database Using a Single Training Sample per Person
  • Table2: Comparative Error Rates of SRC and ESRC on the AR Database Using a Single Training Sample per Person represent the expressional variations. In general, our ESRC methods provide a novel and unified solution on the four variabilities
  • Table3: Comparative Recognition Rates of SRC and ESRC on the FERET Database Using the FERET’96 Testing Protocol years. The image is first normalized by an affine transformation that sets the centered intereye line horizontal and 70 pixel apart, and then cropped to the size of 128 Â 128 with the centers of the eyes4 located at (29, 34) and (99, 34) to extract the pure face region. No further preprocessing procedure is carried out in our experiments, and Fig. 8 shows some cropped images which are used in our experiments
Download tables as Excel
基金
  • This work was supported by the National Natural Science Foundation of China (NSFC) under Grant Nos. 61002051 and 61005025, Major Project of National Science and Technology under Grant No 2011ZX03002-005-01, and the Fundamental Research Funds for the Central Universities
  • For more information on this or any other computing topic, please visit our under Grant Nos. 2011RC0102, 2009RC0106, and 2011RC0115. Digital Library at www.computer.org/publications/dlib
引用论文
  • P.N. Belhumeour, J.P. Hespanha, and D.J. Kriegman, “Eigenfaces versus Fisherfaces: Recognition Using Class Specific Linear Projection,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 19, no. 7, pp. 711-720, July 1997.
    Google ScholarLocate open access versionFindings
  • D.L. Swets and J.J. Weng, “Using Discriminant Eigenfeatures for Image Retrieval,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 18, no. 8, pp. 831-836, Aug. 1996.
    Google ScholarLocate open access versionFindings
  • C. Liu and H. Wechsler, “Gabor Feature Based Classification Using the Enhanced Fisher Linear Discriminant Model for Face Recognition,” IEEE Trans. Image Processing, vol. 11, no. 4, pp. 467-476, Apr. 2002.
    Google ScholarLocate open access versionFindings
  • X. He, S. Yan, Y. Hu, P. Niyogi, and H.-J. Zhang, “Face Recognition Using Laplacianfaces,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 27, no. 3, pp. 328-340, Mar. 2005.
    Google ScholarLocate open access versionFindings
  • X. He and P. Niyogi, “Locality Preserving Projections,” Proc. Conf. Advances in Neural Information Processing System, 2003.
    Google ScholarLocate open access versionFindings
  • W. Deng, J. Hu, and J. Guo, “Comments on ‘Globally Maximizing, Locally Minimizing: Unsupervised Discriminant Projection with Applications to Face and Palm Biometrics’,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 30, no. 8, pp. 1503-1504, Aug. 2008.
    Google ScholarLocate open access versionFindings
  • J. Wright, A. Yang, A. Ganesh, S. Sastry, and Y. Ma, “Robust Face Recognition via Sparse Representation,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 31, no. 2, pp. 210-227, Feb. 2009.
    Google ScholarLocate open access versionFindings
  • A. Wagner, J. Wright, A. Ganesh, Z. Zhou, and Y. Ma, “Towards a Practical Face Recognition System: Robust Registration and Illumination by Sparse Representation,” Proc. IEEE Conf. Computer Vision and Pattern Recognition, pp. 597-604, 2009.
    Google ScholarLocate open access versionFindings
  • A.M. Martinez and R. Benavente, “The AR Face Database,” CVC Technical Report #24, June 1998.
    Google ScholarFindings
  • P.J. Phillips, H. Moon, P. Rizvi, and P. Rauss, “The Feret Evaluation Method for Face Recognition Algorithms,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 22, no. 10, pp. 1090-1104, Oct. 2000.
    Google ScholarLocate open access versionFindings
  • D. Donoho and Y. Tsaig, “Fast Solution of ‘1-Norm Minimization Problems When the Solution May Be Sparse,” IEEE Trans. Information Theory, vol. 54, no. 11, pp. 4789-4812, Nov. 2008.
    Google ScholarLocate open access versionFindings
  • A. Georghiades, P. Belhumeur, and D. Kriegman, “From Few to Many: Illumination Cone Models for Face Recognition under Variable Lighting and Pose,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 23, no. 6, pp. 643-660, June 2001.
    Google ScholarLocate open access versionFindings
  • A. Wagner, J. Wright, A. Ganesh, Z. Zhou, H. Mobahi, and Y. Ma, “Toward a Practical Face Recognition System: Robust Alignment and Illumination by Sparse Representation,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 34, no. 2, pp. 372-386, Feb. 2012.
    Google ScholarLocate open access versionFindings
  • A. Yang, A. Ganesh, S. Sastry, and Y. Ma, “Fast l1-Minimization Algorithms and an Application in Robust Face Recognition: A Review,” Technical Report UCB/EECS-2010-13, Univ. of California, 2010.
    Google ScholarFindings
  • M. Osborne, B. Presnell, and B. Turlach, “A New Approach to Variable Selection in Least Squares Problems,” IMA J. Numerical Analysis, vol. 20, no. 3, pp. 389-403, 2000.
    Google ScholarLocate open access versionFindings
  • W. Deng, J. Hu, J. Guo, W. Cai, and D. Feng, “Emulating Biological Strategies for Uncontrolled Face Recognition,” Pattern Recognition, vol. 43, no. 6, pp. 2210-2223, 2010.
    Google ScholarLocate open access versionFindings
  • W. Deng, J. Hu, J. Guo, W. Cai, and D. Feng, “Robust, Accurate and Efficient Face Recognition from a Single Training Image: A Uniform Pursuit Approach,” Pattern Recognition, vol. 43, no. 5, pp. 1748-1762, 2010.
    Google ScholarLocate open access versionFindings
  • C. Chan and J. Kittler, “Sparse Representation of (Multiscale) Histograms for Face Recognition Robust to Registration and Illumination Problems,” Proc. IEEE 17th Int’l Conf. Image Processing, pp. 2441-2444, 2010.
    Google ScholarLocate open access versionFindings
  • M. Elad and M. Aharon, “Image Denoising via Sparse and Redundant Representations over Learned Dictionaries,” IEEE Trans. Image Processing, vol. 15, no. 12, pp. 3736-3745, Dec. 2006.
    Google ScholarLocate open access versionFindings
  • W. Deng, J. Guo, and J. Hu, “Comment on ‘100 Percent Accuracy in Automatic Face Recognition’,” Science, vol. 321, no. 5891, p. 912, 2008.
    Google ScholarLocate open access versionFindings
0
您的评分 :

暂无评分

标签
评论
数据免责声明
页面数据均来自互联网公开来源、合作出版商和通过AI技术自动分析结果,我们不对页面数据的有效性、准确性、正确性、可靠性、完整性和及时性做出任何承诺和保证。若有疑问,可以通过电子邮件方式联系我们:report@aminer.cn