pymoten: motion energy features from video using a pyramid of spatio-temporal Gabor filters
Authors/Creators
- 1. University of California, Berkeley
Description
pymoten is a set of tools to measure local pixel motion in videos. The motion energy features are computed from the video using a pyramid of spatio-temporal Gabor filters. The spatio-temporal Gabor filters are created at multiple spatial and temporal frequencies, directions of motion, x-y positions, and sizes. Each filter quadrature-pair is convolved with the video and their activation energy is computed for each video frame. Empirically, motion energy features provide a good basis to model brain responses to natural movies and to reconstruct natural movies from brain responses [@Nishimoto2011a, @Nishimoto2011b] and is inspired by the way in which neurons in the visual cortex work.