'Measuring how the human visual system combines stereo and blur cues to estimate depth' - COMP 396 Undergraduate Research Project Application Form

Supervisor's Name: Michael Langer

Supervisor's Email: michael.langer [at] mcgill.ca

Supervisor's Phone: 514-398-3740

Supervisor's Website: http://www.cim.mcgill.ca/~langer

Supervisor's department: Computer Science

Course number: COMP 396 (Computer Science)

Term: Fall 2013-2014

Project start date: Tuesday, September 3, 2013

Project end date: Tuesday, December 3, 2013

Project title: Measuring how the human visual system combines stereo and blur cues to estimate depth

Project description (50-100 words suggested): The visual system uses many sources of information to estimate the depth of objects in a scene. Two examples are stereo disparity and optical blur. The project will investigate how these two cues are combined, using an experiment that is similar to one carried out by Held, Cooper, and Banks [Current Biology, 2012]. Images will be synthesized using computer graphics and presented on a 3D stereo display. Human subjects will compare the depths of a pair of surfaces based on disparity and blur cues. (For subjects, we will use grad students in the lab.) We will measure the accuracy with which subjects can perform the depth discrimination task for the blue and stereo disparity cues on their own, and in combination.

Prerequisite: 1 term completed at McGill + CGPA of 3.0 or higher; or permission of instructor.

Grading scheme (The final report must be worth at least 50% of final grade): Final grade will be 50% lab performance (meeting timetable/workload plan that is specified below) and the Final Report (50%) which is due on Dec. 3, 2013. The Final Report should be at least 2000 words. It should contain four sections: Introduction, Methods, Results, Discussion. The experiment will be similar to that done by Held et al (2012), so similar details can be replaced to references to that paper. More important will be to discuss any differences.

Other project information: The Timetable and Reading List are given below. The workload will be about 40 hours per month, broken down as follows:

September:

  • (20 hours) Read journal papers in Reading List below, and discuss with supervisor
  • (10 hours) Learn basics of python programming language e.g. http://www.codecademy.com/
  • (10 hours) Install PsychoPy application http://www.psychopy.org/ and implement the "stroop test" experiment, discussed in youtube tutorial. Using an anaglyph display and within PsychoPy, set up a basic stereo and focus stimulus such as was used in the paper by Held, Cooper, Banks in Current Biology 2012.

October:

  • (10 hours) Read chapters 1-4 of Psychophysics textbook by Kingdom and Prins
  • (30 hours) Figure out how to present the stimuli on the 3D stereo display in the lab (i.e. using shutter glasses rather than anaglyph), and within the PsychoPy application. Using PsychoPy, design an experiment to measure the just-noticable-difference for blur and stereo disparity, similar to what was done in Held, Cooper, Banks 2012 (Fig. 2).

November:

  • (20 hours) Run the experiment using several subjects (including at least two naive subjects). Calculate JNDs, plot data.
  • (20 hours) Write the Final Report.

Reading List:

  • F. Kingdom and N. Prins, Psychophysics: a practical introduction (2009) Ch. 1-4
  • Watt, S. J., Akeley, K., Ernst, M. O., & Banks, M. S. (2005). Focus cues affect perceived depth. Journal of Vision, 5, 834-862.
  • Watson AB and Ahumada AJ, "Blur clarified: a review and synthesis of blur discrimination" Journal of Vision (2011)
  • Held, Cooper, Banks "Blur and disparity are complementary cues to depth", Current Biology 22: 426-431, 2012
  • Hoffman DM, Banks MS, "Focus information is used to interpret binocular images", Journal of Vision 10(5), 2010

The following two papers should be available from http://www.georgemather.com/Publications.html

  • Mather G, Smith D (2002) Blur discrimination and its relation to blur-mediated depth perception. Perception, 31, 1211-1219.
  • Mather G, Smith D (2000) Depth cue integration: stereopsis and image blur. Vision Research, 40, 3501-3506.

Project status: This project is taken. The professor has no more '396' projects this term.

Ethics, safety, and training: Supervisors are responsible for the ethics and safety compliance of undergraduate students. This project involves human subjects.