Domain adaption of vehicle detector based on convolutional neural networks

Regular Papers Intelligent and Information Systems

Abstract

Generally the performance of a vehicle detector will decrease rapidly, when it is trained on a fixed training set but applied to a specific scene with view changes. The reason is that in the training set only a few samples are helpful for vehicle detection in the specific scene while other samples disturb the accurate detections. To solve this problem, we propose a novel transfer learning method to adapt the trained vehicle detector based on convolutional neural networks (ConvNets) to a specific scene with several new labeled samples. At first we reserve the share-filters and update the non-shared filters to improve the sensitivity of the vehicles in the specific scene. Then we combine the similar feature maps to accelerate the detection speed. At last for making the vehicle detector stable, we fine-tune it several times with the updated training set. Our contributions are an original research on transferring the vehicle detector based on ConvNets and an optimization approach about removing the redundant connections in the ConvNet vehicle detector. The extensive comparative experiments on three different datasets demonstrate that the transferred detectors achieve the improvements on both of the accuracy and speed.

Keywords

Convolutional neural networks domain adaption transfer learning vehicle detection 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. [1]
    T. H. Chen, Y. F. Lin, and T. Y. Chen, “Intelligent vehicle counting method based on blob analysis in traffic surveillance,” Proc. of IEEE Innovative Computing, Information and Control, pp. 238–238, 2007.Google Scholar
  2. [2]
    X. Pan, Y. Guo, and A. Men, “Traffic surveillance system for vehicle flow detection,” Proc. of IEEE Computer Modeling and Simulation, pp. 314–318, 2010.Google Scholar
  3. [3]
    F. Han, Y. Shan, R. Cekander, H. S. Sawhney, and R. Kumar, “A two-stage approach to people and vehicle detection with hog-based SVM,” Proc. of Performance Metrics for Intelligent Systems 2006 Workshop, pp. 133–140, 2006.Google Scholar
  4. [4]
    H. Cheng, N. Zheng, and C. Sun, “Boosted Gabor features applied to vehicle detection,” Proc. of IEEE International Conference on Pattern Recognition, pp. 662–666, 2006.Google Scholar
  5. [5]
    W. Zheng and L. Liang, “Fast car detection using image strip features,” Proc. of IEEE Conference on Computer Vision and Pattern Recognition, pp. 2703–2710, 2009.Google Scholar
  6. [6]
    X. Jin and C. H. Davis, “Vehicle detection from high-resolution satellite imagery using morphological shared-weight neural networks,” Image and Vision Computing, vol. 25, no. 9, pp. 1422–1431, 2007.CrossRefGoogle Scholar
  7. [7]
    B. Wu and R. Nevatia, “Cluster boosted tree classifier for multi-view, multi-pose object detection,” Proc. of IEEE International Conference on Computer Vision, pp. 1–8, 2007.Google Scholar
  8. [8]
    H. Y. Cheng, C. C. Weng, and Y. Y. Chen, “Vehicle detection in aerial surveillance using dynamic Bayesian networks,” IEEE Trans. on Image Processing, vol. 21, no. 4, pp. 2152–2159, 2012.MathSciNetCrossRefGoogle Scholar
  9. [9]
    N. Dalal and B. Triggs, “Histograms of oriented gradients for human detection,” Proc. of IEEE Conference on Computer Vision and Pattern Recognition, pp. 886–893, 2005.Google Scholar
  10. [10]
    C. Huang, H. Ai, Y. Li, and S. Lao, “Vector boosting for rotation invariant multi-view face detection,” Proc. of IEEE International Conference on Computer Vision, pp. 446–453, 2005.Google Scholar
  11. [11]
    G. E. Hinton and R. R. Salakhutdinov, “Reducing the dimensionality of data with neural networks,” Science, vol. 313, no. 5786, pp. 504–507, 2006.MathSciNetCrossRefMATHGoogle Scholar
  12. [12]
    Y. Bengio, P. Lamblin, D. Popovici, and H. Larochelle, “Greedy layer-wise training of deep networks,” Proc. of Advances in Neural Information Processing Systems, pp. 153, 2007.Google Scholar
  13. [13]
    P. Xu, M. Ye, Q. Liu, X. Li, L. Pei, and J. Ding, “Motion detection via a couple of auto-encoder networks,” Proc. of IEEE International Conference on Multimedia and Expo, pp. 1–6, 2014.Google Scholar
  14. [14]
    Y. LeCun, L. Bottou, Y. Bengio, and P. Haffner, “Gradient-based learning applied to document recognition,” Proc. of the IEEE, pp. 2278–2324, 1998.Google Scholar
  15. [15]
    A. Krizhevsky, I. Sutskever, and G. E. Hinton, “ImageNet Classification with Deep Convolutional Neural Networks,” Proc. of Advances in Neural Information Processing Systems, pp. 4, 2012.Google Scholar
  16. [16]
    C. Garcia and M. Delakis, “Convolutional face finder: A neural architecture for fast and robust face detection,” IEEE Trans. on Pattern Analysis and Machine Intelligence, vol. 26, no. 11, pp. 1408–1423, 2004.CrossRefGoogle Scholar
  17. [17]
    Y. Chen, C. Han, C. Wang, B. Jeng, and K. Fan, “A CNN-based face detector with a simple feature map and a coarse-to-fine classifier,” IEEE Trans. on Pattern Analysis and Machine Intelligence, vol. PP, no. 99, pp. 1–13, 2010.Google Scholar
  18. [18]
    P. Sermanet, K. Kavukcuoglu, S. Chintala, and Y. LeCun, “Pedestrian detection with unsupervised multi-stage feature learning,” Proc. of IEEE Conference on Computer Vision and Pattern Recognition, pp. 3626–3633, 2013.Google Scholar
  19. [19]
    J. Pang, Q. Huang, S. Yan, S. Jiang, and L. Qin, “Transferring boosted detectors towards viewpoint and scene adaptiveness,” IEEE Trans. on Image Processing, vol. 20, no. 5, pp. 1388–1400, 2011.MathSciNetCrossRefGoogle Scholar
  20. [20]
    M. Wang, W. Li, and X. Wang, “Transferring a generic pedestrian detector towards specific scenes,” Proc. of IEEE Conference on Computer Vision and Pattern Recognition, pp. 3274–3281, 2012.Google Scholar
  21. [21]
    K. Jarrett, K. Kavukcuoglu, M. A. Ranzato, and Y. LeCun, “What is the best multi-stage architecture for object recognition?,” Proc. of IEEE International Conference on Computer Vision, pp. 2146–2153, 2009.Google Scholar
  22. [22]
    P. Viola and M. Jones, “Rapid object detection using a boosted cascade of simple features,” Proc. of IEEE Conference on Computer Vision and Pattern Recognition, pp. 511–518, 2001.Google Scholar
  23. [23]
    H. J. Seo and P. Milanfar, “Training-free, generic object detection using locally adaptive regression kernels,” IEEE Trans. on Pattern Analysis and Machine Intelligence, vol. 32, no. 9, pp. 1688–1704, 2010.CrossRefGoogle Scholar
  24. [24]
    T. L. Saaty, Analytic Hierarchy Process, Springer, 2013.CrossRefGoogle Scholar
  25. [25]
    F. F. Li, R. Fergus, and P. Perona, “One-shot learning of object categories,” IEEE Trans. on Pattern Analysis and Machine Intelligence, vol. 28, no. 4, pp. 594–611, 2006.CrossRefGoogle Scholar
  26. [26]
    Caltech computational vision Caltech cars 1999. [Online]. Available: http://www.vision.caltech.edu/ html-files/archive.html.Google Scholar
  27. [27]
    S. Agarwal, A. Awan, and D. Roth, “Learning to detect objects in images via a sparse, part-based representation,” IEEE Trans. on Pattern Analysis and Machine Intelligence, vol. 26, no. 11, pp. 1475–1490, 2004.CrossRefGoogle Scholar
  28. [28]
    X. Wang, X. Ma, and W. E. L. Grimson, “Unsupervised activity perception in crowded and complicated scenes using hierarchical Bayesian models,” IEEE Trans. on Pattern Analysis and Machine Intelligence, vol. 31, no. 3, pp. 539–555, 2009.CrossRefGoogle Scholar
  29. [29]
    P. Carbonetto, G. Dorkó, C. Schmid, H. Kück, and N. De Freitas, “Learning to recognize objects with little supervision,” International Journal of Computer Vision, vol. 77, no. 1-3, pp. 219–237, 2008.CrossRefGoogle Scholar
  30. [30]
    P. Xu, M. Ye, X. Li, L. Pei, and P. Jiao, “Object detection using voting spaces trained by few samples,” Optical Engineering, vol. 52, no. 9, pp. 093105-093105, 2013.Google Scholar
  31. [31]
    X. Yang, H. Liu, and L. J. Latecki, “Contour-based object detection as dominant set computation,” Pattern Recognition, vol. 45, no. 5, pp. 1927–1936, 2012.CrossRefGoogle Scholar
  32. [32]
    T. Li, M. Ye, and J. Ding, “Discriminative Hough context model for object detection,” The Visual Computer, vol. 30, no. 1, pp. 59–69, 2014.CrossRefGoogle Scholar
  33. [33]
    C.-C. R. Wang and J.-J. Lien, “Automatic vehicle detection using local features—a statistical approach,” IEEE Trans. on Intelligent Transportation Systems, vol. 9, no. 1, pp. 83–96, 2008.CrossRefGoogle Scholar
  34. [34]
    A. Kapoor and J. Winn, “Located hidden random fields: Learning discriminative parts for object detection,” Proc. of European Conference on Computer Vision, pp. 302–315, 2006.Google Scholar
  35. [35]
    B. Wu and R. Nevatia, “Simultaneous object detection and segmentation by boosting local shape feature based classifier,” Proc. of IEEE Conference on Computer Vision and Pattern Recognition, pp. 1–8, 2007.Google Scholar
  36. [36]
    T. Li, M. Ye, F. Pang, H. Y. Wang, and J. Ding, “An efficient fire detection method based on orientation feature,” International Journal of Control, Automation, and Systems, vol. 11, no. 5, pp. 1038–1045, 2013.CrossRefGoogle Scholar

Copyright information

© Institute of Control, Robotics and Systems and The Korean Institute of Electrical Engineers and Springer-Verlag Berlin Heidelberg 2015

Authors and Affiliations

  1. 1.School of Computer Science and Engineering, Center for Robotics and Key Laboratory for Neuro Information of Ministry of EducationUniversity of Electronic Science and Technology of ChinaChengduP. R. China

Personalised recommendations