Computer Vision

2014 Edition
| Editors: Katsushi Ikeuchi

Weakly Supervised Learning

  • Lorenzo Torresani
Reference work entry



Weakly supervised learning is a machine learning framework where the model is trained using examples that are only partially annotated or labeled.


Most modern computer vision system involves models learned from human-labeled image examples. For instance, an object detector is typically trained on a large collection of images manually annotated with masks or bounding boxes denoting the location of the object of interest in each photo. The reliance on time-consuming human labeling poses a significant limitation to the practical application of these methods. Weakly supervised learning is aimed at reducing the amount of human intervention needed to train the models by making use of examples that are only partially labeled.


There are two main forms of weakly supervised learning, differing with respect to the type of partial labels used to annotate the examples:
  1. 1.

    Semisupervised learninginvolves training a model using a...

This is a preview of subscription content, log in to check access.


  1. 1.
    . Babenko B, Ming-Hsuan Yang, Belongie S (2009) Visual tracking with online multiple instance learning. In: IEEE conference on computer vision and pattern recognition (CVPR), Miami, pp 983–990Google Scholar
  2. 2.
    . Blake A, Rother C, Brown M, Pérez P, Torr PHS (2004) Interactive image segmentation using an adaptive gmmrf model. In: 8th European conference on computer vision (ECCV), Prague, pp 428–441Google Scholar
  3. 3.
    . Blum A, Mitchell TM (1998) Combining labeled and unlabeled data with co-training. In: Computational learning theory, Madison, Wisconsin, USA, pp 92–100Google Scholar
  4. 4.
    . Deselaers T, Alexe B, Ferrari V (2012) Weakly supervised localization and learning with generic knowledge. Int J Comp Vis 1–19Google Scholar
  5. 5.
    Dietterich TG, Lathrop RH, Lozano-Pérez T (1997) Solving the multiple instance problem with axis-parallel rectangles. Artif Intell 89(1–2):31–71CrossRefzbMATHGoogle Scholar
  6. 6.
    Felzenszwalb PF, Girshick RB, McAllester DA, Ramanan D (2010) Object detection with discriminatively trained part-based models. IEEE Trans Pattern Anal Mach Intell 32(9):1627–1645CrossRefGoogle Scholar
  7. 7.
    . Fergus R, Perona P, Zisserman A (2004) A visual category filter for google images. In: ECCV 2004, 8th European conference on computer vision, Prague, pp 242–256Google Scholar
  8. 8.
    . Fergus R, Weiss Y, Torralba A (2009) Semi-supervised learning in gigantic image collections. In: Advances in neural information processing systems 22, Vancouver, pp 522–530Google Scholar
  9. 9.
    Grabner H, Leistner C, Bischof H (2008) Semi-supervised on-line boosting for robust tracking. In: Forsyth D, Torr P, Zisserman A (eds) Computer vision (ECCV 2008). Lecture notes in computer science, vol 5302. Springer, Berlin/Heidelberg, pp 234–247Google Scholar
  10. 10.
    Levin A, Lischinski D, Weiss Y (2004) Colorization using optimization. ACM Trans Graph 23(3):689–694CrossRefGoogle Scholar
  11. 11.
    . Levin A, Viola PA, Freund Y (2003) Unsupervised improvement of visual detectors using co-training. In: 9th IEEE international conference on computer vision (ICCV), Nice, pp 626–633Google Scholar
  12. 12.
    . Navaratnam R, Fitzgibbon AW, Cipolla R (2007) The joint manifold model for semi-supervised multi-valued regression. In: IEEE 11th international conference on computer vision (ICCV), Rio de Janeiro, pp 1–8Google Scholar
  13. 13.
    . Nguyen MH, Torresani L, Lorenzo de la Torre, Rother C (2009) Weakly supervised discriminative localization and classification: a joint learning process. In: IEEE 12th international conference on computer vision (ICCV), Kyoto, pp 1925–1932Google Scholar
  14. 14.
    . Nigam K, Ghani R (2000) Analyzing the effectiveness and applicability of co-training. In: ACM CIKM international conference on information and knowledge management, McLean, VA, USA, pp 86–93Google Scholar
  15. 15.
    . Rosenberg C, Hebert M, Schneiderman H (2005) Semi-supervised self-training of object detection models. In: 7th IEEE workshop on applications of computer vision/IEEE workshop on motion and video computing, Breckenridge, CO, USA, pp 29–36Google Scholar
  16. 16.
    Zhu X, Goldberg AB (2009) Introduction to semi-supervised learning. Synthesis lectures on artificial intelligence and machine learning. Morgan & Claypool, San Rafael/CaliforniaGoogle Scholar

Copyright information

© Springer Science+Business Media New York 2014

Authors and Affiliations

  • Lorenzo Torresani
    • 1
  1. 1.Computer Science Department, Dartmouth CollegeHanover, NHUSA