Advertisement

Machine Vision and Applications

, Volume 25, Issue 3, pp 599–611 | Cite as

Active learning for on-road vehicle detection: a comparative study

  • Sayanan Sivaraman
  • Mohan M. Trivedi
Special Issue Paper

Abstract

In recent years, active learning has emerged as a powerful tool in building robust systems for object detection using computer vision. Indeed, active learning approaches to on-road vehicle detection have achieved impressive results. While active learning approaches for object detection have been explored and presented in the literature, few studies have been performed to comparatively assess costs and merits. In this study, we provide a cost-sensitive analysis of three popular active learning methods for on-road vehicle detection. The generality of active learning findings is demonstrated via learning experiments performed with detectors based on histogram of oriented gradient features and SVM classification (HOG–SVM), and Haar-like features and Adaboost classification (Haar–Adaboost). Experimental evaluation has been performed on static images and real-world on-road vehicle datasets. Learning approaches are assessed in terms of the time spent annotating, data required, recall, and precision.

Keywords

Semi-supervised learning Active learning Annotation costs Object detection Active safety Intelligent vehicles 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Abramson, Y., Freund, Y.: Active learning for visual object detection. UCSD Technical Report (2006). http://www.cse.ucsd.edu/~yfreund/papers/sevillePaper.pdf
  2. 2.
    Chan, Y., Huang, S., Fu, L., Hsiao, P.: Vehicle detection under various lighting conditions by incorporating particle filter. IEEE Intell. Transp. Syst. Conf. (2007)Google Scholar
  3. 3.
    Chang, C.C., Lin, C.J.: LIBSVM: a library for support vector machines. ACM Trans. Intell. Syst. Technol. 2(3) (2011) (Software available at http://www.csie.ntu.edu.tw/~cjlin/libsvm)
  4. 4.
    Cohn D., Atlas L., Ladner R.: Improving generalization with active learning. Mach. Learn. 15(2), 201–221 (1994)Google Scholar
  5. 5.
    Collins, M., Schapire, R.E., Singer, Y.: Logistic regression, Adaboost and Bregman distances. Mach. Learn. (2002)Google Scholar
  6. 6.
    Cortes C., Vapnik V.: Support vector networks. Mach. Learn. 20, 273–297 (1995)zbMATHGoogle Scholar
  7. 7.
    Dalal, N., Triggs, B.: Histograms of oriented gradients for human detection. IEEE Conf. Comp. Vis. Patt. Recog. (2005)Google Scholar
  8. 8.
    Dasgupta, S.: Analysis of a greedy active learning strategy. Neural Inf. Process. Syst. (2004)Google Scholar
  9. 9.
    Dasgupta, S., Hsu, D.J., Monteleoni, C.: A general agnostic active learning algorithm. Neural Inf. Process. Syst. (2007)Google Scholar
  10. 10.
    Enzweiler, M., Gavrila, D.M.: A mixed generative-discriminative framework for pedestrian classification. IEEE Comput. Vis. Pattern Recognit. Conf. (2008)Google Scholar
  11. 11.
    Murphy-Chutorian, E., Trivedi, M.M.: Head pose estimation and augmented reality tracking: an integrated system and evaluation for monitoring driver awareness. IEEE Trans. Intell. Transp. Syst. (2010)Google Scholar
  12. 12.
    Fossati, A., Schnmann, P., Fua, P.: Real-time vehicle tracking for driving assistance. Mach. Vis. Appl. (2010)Google Scholar
  13. 13.
    Freund Y., Schapire R.E.: A short introduction to boosting. J. Jpn. Soc. Artif. Intell. 14(5), 771–780 (1999)Google Scholar
  14. 14.
    Friedman, J., Hastie, T., Tibshirani, R.: Additive logistic regression, a statistical view of boosting. Ann. Stat. (2000)Google Scholar
  15. 15.
    Gandhi, T., Trivedi, M.M.: Parametric ego-motion estimation for vehicle surround analysis using an omnidirectional camera. Mach. Vis. Appl. (2005)Google Scholar
  16. 16.
    Grabner H., Bischof H.: On-line boosting and vision. IEEE Comput. Vis. Pattern Recognit. Conf. 1, 260–267 (2006)Google Scholar
  17. 17.
    Haselhoff, A., Schauland, S., Kummert, A.: A signal theoretic approach to measure the influence of image resolution for appearance-based vehicle detection. IEEE Intell. Veh. Symp. (2008)Google Scholar
  18. 18.
    Haselhoff, A., Kummert, A.: An evolutionary optimized vehicle tracker in collaboration with a detection system. IEEE Intell. Transp. Syst. Conf. (2009)Google Scholar
  19. 19.
    Haselhoff, A., Kummert, A.: A vehicle detection system based on Haar and triangle features. IEEE Intell. Veh. Symp. (2009)Google Scholar
  20. 20.
    Hewitt, R., Belongie, S.: Active learning in face recognition, using tracking to build a face model. IEEE Conf. Comp. Vis. Pattern. Recognit. Workshop (2006)Google Scholar
  21. 21.
    Joshi, A.J., Porikli, F.: Scene adaptive human detection with incremental active learning. IAPR Int’l Conf. Pattern. Recognit. (2010)Google Scholar
  22. 22.
    Kapoor, A., Grauman, K., Urtasun, R., Darrell, T.: Gaussian processes for object categorization. Int. J. Comput. Vis. (2009)Google Scholar
  23. 23.
    Krotosky, S.J., Trivedi, M.M.: On color-, infrared-, and multimodal-stereo approaches to pedestrian detection. IEEE Trans. Intell. Transp. Syst. (2007)Google Scholar
  24. 24.
    Lampert, C.H., Peters, J.: Active structured learning for high-speed object detection. DAGM Symposium on Pattern Recognition (2009)Google Scholar
  25. 25.
    Leistner, C., Roth, P., Grabner, H., Bischof, H., Stratzer, A., Rinner, B.: Visual on-line learning in distributed camera networks. IEEE Int Conf. Distrib. Smart Cameras (2008)Google Scholar
  26. 26.
    Li M., Sethi I.K.: Confidence-based active learning. IEEE Trans. Pattern. Anal. Mach. Intell. 28(8), 1251–1261 (2006)CrossRefGoogle Scholar
  27. 27.
    Lin H.-T., Lin C.-J., Weng R.C.: A note on Platts probabilistic outputs for support vector machines. Mach. Learn. 68, 267–276 (2007)CrossRefGoogle Scholar
  28. 28.
    McCall, J., Trivedi, M.M.: Video based lane estimation and tracking for driver assistance: survey, system, and evaluation. IEEE Trans. Intell. Transp. Syst. vol. 7, no. 1 (2006)Google Scholar
  29. 29.
    National Highway Traffic Safety Administration. Distracted Driving 2009 (2010). http://www-nrd.nhtsa.dot.gov/Pubs/811379.pdf
  30. 30.
    Nieto, M., Laborda, J.A., Salgado, L.: Road environment modeling using robust perspective analysis and recursive Bayesian segmentation. Mach. Vis. Appl. (2010)Google Scholar
  31. 31.
    Ponsa, D., Lopez, A., Lumbreras, F., Serrat, J., Graf, T.: 3D vehicle sensor based on monocular vision. IEEE Intell. Transp. Syst. Conf., pp. 1096–1101 (2005)Google Scholar
  32. 32.
    Roth, P.M., Bischof, H.: Active sampling via tracking. IEEE Comput. Vis. Pattern Recognit. Conf. (2008)Google Scholar
  33. 33.
    Roth, P., Grabner, H., Leistner, C., Winter, M., Bischof, H.: Interactive learning a person detector: fewer clicks—less frustration. Workshop of the Austrian Association for Pattern Recognition (AAPR) (2008)Google Scholar
  34. 34.
    Roth, P., Sternig, S., Grabner, H., Bischof, H.: Classifier grids for robust adaptive object detection. IEEE Conf. Comput. Vis. Pattern. Recognit. (2009)Google Scholar
  35. 35.
    Settles, B., Craven, M., Friedland, L.: Active learning with real annotation costs. Nips Workshop on Cost-Sensitive Learning (2008)Google Scholar
  36. 36.
    Sivaraman, S., Trivedi, M.M.: A general active learning framework for on-road vehicle recognition and tracking. IEEE Trans. Intell Transp. Syst. (2010)Google Scholar
  37. 37.
    Sivaraman, S., Trivedi, M.M.: Improved vision-based lane tracker performance using vehicle localization. IEEE Intell. Veh. Symp. (2010)Google Scholar
  38. 38.
    Sivaraman, S., Trivedi, M.M.: Combining monocular and stereo-vision for real-time vehicle ranging and tracking on multilane highways. IEEE Intell. Transp. Syst. Conf. (2011)Google Scholar
  39. 39.
    Sun Z., Bebis G., Miller R.: Monocular Precrash Vehicle Detection: Features and Classifiers. IEEE Trans. Image Proc. 15(7), 2019–2034 (2006)CrossRefGoogle Scholar
  40. 40.
    Sun, Z., Bebis, G., Miller, R.: On-road vehicle detection: a review. IEEE Trans. Pattern Anal. Mach. Intell. (2006)Google Scholar
  41. 41.
    Takeuchi, A., Mita, S., McAllester, D.: On-road vehicle tracking using deformable object model and particle filter with integrated likelihoods. IEEE Intell. Veh. Symp. (2010)Google Scholar
  42. 42.
    Trivedi, M.M., Gandhi, T., McCall, J.: Looking-in and looking-out of a vehicle: computer-vision-based enhanced vehicle safety. IEEE Trans. Intell. Transp. Syst. (2007)Google Scholar
  43. 43.
    Tsuruoka, Y., Tsujii, J., Ananiadou, S.: Accelerating the annotation of sparse named entities by dynamic sentence selection. BioNLP ACL workshop: themes in biomedical language processing (2008)Google Scholar
  44. 44.
    Vijayanarasimhan, S., Grauman, K.: Multi-level active prediction of useful image annotations for recognition. Adv. Neural Inf. Process. Syst. (2008)Google Scholar
  45. 45.
    Viola, P., Jones, M.: Rapid object detection using a boosted cascade of simple features. IEEE Conf. Comput. Vis. Pattern. Recognit., vol. 1, pp. 511–518 (2001)Google Scholar
  46. 46.
    World report on road traffic injury prevention World Health Organization. http://www.who.int/violence_injury_prevention/publications/road_traffic/world_report/factsheets/en/index.html

Copyright information

© Springer-Verlag 2011

Authors and Affiliations

  1. 1.Computer Vision and Robotics Research LabUniversity of California, San DiegoLa JollaUSA

Personalised recommendations