Perf Guidelines

[transitive-bullshit]

I'm trying to use tracking.js with getUserMedia and a face object tracker and am experiencing unusable framerates.

The source video is 640x480, and I'm using the following tracker options:

      tracker.setInitialScale(4)
      tracker.setStepSize(2)
      tracker.setEdgesDensity(0.1)

My question is: can any of the tracker options be used to affect perf vs accuracy, or is the largest factor affecting perf the size of the source video?

[eduardolundgren]

Those options can improve performance, specially the step size and initial scale. ViolaJones is a good object detection algorithm, although for large images 640x480 it can reach low frame rates.

Two features from our roadmap that will speed up performance for cases like that:
1- Create some sort of ThreadableTracker abstract class that splits process into different Workers.
2- Use optical flow to predict the movement in order to not require processing every frame using ViolaJones.

A workaround for now is to write the 640x480 video into a smaller hidden canvas, and this canvas is the element being track, when track event fires with rectangles you can proportionally projects the coordinates into the larger visible video.

[transitive-bullshit]

Thanks for the quick response. For now, I'm going with the intermediate canvas approach 👍

[eduardolundgren]

Sounds good, closing this for now.

[eduardolundgren]

Relates to #99.