Research

Scenes and views

Scenes, like objects, are three-dimensional, but we experience them as primarily 2-dimensional views. What makes a "canonical” view of scene? Are some views of scenes more typical than others?

  • Xiao, J., Ehinger, K. A., Oliva, A., & Torralba, A. (2012). Recognizing scene viewpoint using panoramic place representation. Proc. 25th IEEE Conference on Computer Vision and Pattern Recognition (CVPR). [website] [pdf]
  • Ehinger, K. A., Xiao, J., Torralba, A., & Oliva, A. (2011). Estimating scene typicality from human ratings and image features. In L. Carlson, C. Hölscher, & T. Shipley (Eds.), Proceedings of the 33rd Annual Conference of the Cognitive Science Society (pp. 2562-2567). Austin, TX: Cognitive Science Society. [pdf] [slides]
  • Ehinger, K. A., & Oliva, A. (2011). Canonical views of scenes depend on the shape of the space. In L. Carlson, C. Hölscher, & T. Shipley (Eds.), Proceedings of the 33rd Annual Conference of the Cognitive Science Society (pp. 2114-2119). Austin, TX: Cognitive Science Society. [pdf]

Understanding where people look in scenes

I use computer modelling to investigate how image features guide people's eye movements when searching for objects or free-viewing natural scenes.

  • Xu, P., Ehinger, K. A, Zhang, Y., Finkelstein, A., Kulkarni, S. R., & Jianxiong, X. (2015). TurkerGaze: Crowdsourcing saliency with webcam based eye tracking. [website] [pdf]
  • Ehinger, K. A., Hidalgo-Sotelo, B., Torralba, A., & Oliva, A. (2009). Modeling search for people in 900 Scenes: A combined source model of eye guidance. Visual Cognition, 17, 945-978. [website] [pdf]
  • Judd, T., Ehinger, K., Durand, F., Torralba, A. (2009). Learning to predict where people look. In 12th IEEE International Conference on Computer Vision (ICCV), 2106-2113. [website] [pdf]