VSS Analysis Continues
It never ends.
Depth maps.
Pete Experiment
Going to have to hurry up to get my stuff ready for VSS.
So much data, so little time.
A project I’m working on with Julia Mazzeralla and my ol’ pal Pete Docter.
Slippy slidy particles
A friend of mine asked me a question about the way that standing waves make sand on a 2D plate make pretty patterns.
So, I whipped up a quick 1D example in Mathematica.
I threw 50 randomly distributed (in x) particles at a string and let them jump around as a function of the friction on the string and a sort of faux-gravity.
I didn’t get the friction quite right of the particle / surface interaction. Probably messed up putting the energy into the particles too. But, it took 10 minutes and I got this pretty thing (pls forgive the crappy background as GraphicConverter does something weird with the colors when it converts to GIF here).
Recurrence, heat maps, trajectories and fixations
More analysis from Dave Jacobs’ eye tracking project. We wrote all the analysis code ourselves in Mathematica. There’s a bug in the heat map compositing, but I shall procrastinate a solution soon.
Lifted largely from Eye Tracking: A comprehensive guide to methods and measures
We used some ideas from Anderson’s paper on recurrence, modifying them to make a little more sense. We’re still analyzing everything, but we have some interesting results so far.
Recurrence plot
Eye tracking recurrence ala Anderson et al. 2013.
With Dave Jacobs and the amazing Yarbus and Babcock software packages.
Curvature Field Computations
Curvature field calculations for a random noisy shape.
Colors represent magnitude of curvature. Iso-contours also shown in lovely pink.
Hairy Glaven
Hairy elongated noisy stimulus.
The lines are surface normals, sampled across the object. They are generated using a continuous random method (sometimes called ‘noise’) and were used to test symmetry perception in 3D shape.
Sculpted Golden Boy
A stimulus, # 13, otherwise known as ‘golden boy’
Colors indicate mapping of under- and over-sculpt from a shape production experiment where we had artists and non-artists reproduce shapes using only vision, touch, or some combination of both to observe the ‘target’ stimulus. Red indicates where sculptors put too much clay and blue where they didn’t put enough.
Statistical Shape of 3D Objects.
Statistical distribution of shape on a 3D object.
A bell pepper from Norman et al.
Each color represents a class of shape ala Koenderink’s “Shape Index”. The histogram below shows the distribution along with a key to the shape categories illustrated above.
Anticipation of Sabre Fencing Attacks
A set of motion-capture trajectories for the Anticipation of Sabre Fencing Attacks paper by Pete Possidente and I.
You can see the ‘whip’ of the tip of the sabre at the end of the attack (green dots on top). The rest are body joint positions. The whole attack took place over about 3m in about 2s.