A Data-Driven Approach to Understanding Skill in Photographic Composition

  • Todd S. Sachs
  • Ramakrishna Kakarala
  • Shannon L. Castleman
  • Deepu Rajan
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 6469)


Photography requires not only equipment but also skill to reliably produce aesthetically-pleasing results. It can be argued that, for photography, skill is apparent even without sophisticated equipment. However, no scientific tests have been carried out to confirm that supposition. For that matter, there has been little scientific study on whether skill is apparent, whether it can be discerned by judges in blind tests. We report results of an experiment in which 33 subjects were asked to use identical cameras to photograph each of 7 pre-determined scenes, including a portrait, landscapes, and several man-made objects. Each photograph was then rated in a double-blind manner by 8 judges. Of those judges, 3 are professional photographic experts, and 5 are imaging researchers. The results show that expert judges are able to discern photographic skill to a statistically significant level, but that the enthusiasts, who are more akin to the general public, are not. We also analyse the photos using computer vision methods published in the literature, and find that there is no correlation between human judgements and the previously-published machine learning methods.


Skill Level Skill Rating Human Judge Photo Major Computer Vision Method 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Savakis, A., Etz, S., Loui, A.: Evaluation of image appeal in consumer photography. In: SPIE Human Vision and Electronic Imaging V (2000)Google Scholar
  2. 2.
    Tong, H., Li, M., Zhang, H.J., He, J., Zhang, C.: Classification of digital photos taken by photographers or home users. In: Proceedings of Pacific Rim Conference on Multimedia, pp. 198–205. Springer, Heidelberg (2004)Google Scholar
  3. 3.
    Ke, Y., Tang, X., Jing, F.: The design of high-level features for photo quality assessment. In: CVPR (1), pp. 419–426. IEEE Computer Society, Los Alamitos (2006)Google Scholar
  4. 4.
    Datta, R., Joshi, D., Li, J., Wang, J.Z.: Studying aesthetics in photographic images using a computational approach. In: Leonardis, A., Bischof, H., Pinz, A. (eds.) ECCV 2006. LNCS, vol. 3953, pp. 288–301. Springer, Heidelberg (2006)CrossRefGoogle Scholar
  5. 5.
    Datta, R., Wang, J.Z.: Acquine: aesthetic quality inference engine - real-time automatic rating of photo aesthetics. In: Wang, J.Z., Boujemaa, N., Ramirez, N.O., Natsev, A. (eds.) Multimedia Information Retrieval, pp. 421–424. ACM, New York (2010)Google Scholar
  6. 6.
    Luo, Y., Tang, X.: Photo and video quality evaluation: Focusing on the subject. In: Forsyth, D., Torr, P.H.S., Zisserman, A. (eds.) ECCV 2008, Part III. LNCS, vol. 5304, pp. 386–399. Springer, Heidelberg (2008)CrossRefGoogle Scholar
  7. 7.
    Zakia, R.: Perception and imaging: photography–a way of seeing, 3rd edn. Elsevier Science Ltd., Cambridge (2007)Google Scholar
  8. 8.
    Maritz, J.S.: Distribution-free statistical methods. Chapman and Hall, London (1991)zbMATHGoogle Scholar
  9. 9.
    Snavely, N., Seitz, S.M., Szeliski, R.: Photo tourism: Exploring photo collections in 3d. In: SIGGRAPH Conference Proceedings, pp. 835–846. ACM Press, New York (2006)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2011

Authors and Affiliations

  • Todd S. Sachs
    • 1
  • Ramakrishna Kakarala
    • 2
  • Shannon L. Castleman
    • 2
  • Deepu Rajan
    • 2
  1. 1.Aptina ImagingSan JoseUSA
  2. 2.Nanyang Technological UniversitySingapore

Personalised recommendations