mi@0: """
mi@0: Foote method for segmentation, published here:
mi@0: 
mi@0: Foote, J. (2000). Automatic Audio Segmentation Using a Measure Of Audio
mi@0: Novelty. In Proc. of the IEEE International Conference of Multimedia and Expo
mi@0: (pp. 452-455). New York City, NY, USA.
mi@0: """
mi@0: 
mi@0: __author__ = "Oriol Nieto"
mi@0: __copyright__ = "Copyright 2014, Music and Audio Research Lab (MARL)"
mi@0: __license__ = "GPL"
mi@0: __version__ = "1.0"
mi@0: __email__ = "oriol@nyu.edu"
mi@0: 
mi@0: # Local stuff
mi@0: from utils import SegUtil
mi@0: 
mitian@7: M = 2           # Median filter for the audio features (in beats)
mitian@7: Mg = 32         # Gaussian kernel size
mitian@7: L = 16          # Size of the median filter for the adaptive threshold
mi@0: 
mi@0: def segmentation(F, M, Mg, L, plot=False):
mi@0:     """Computes the Foote segmentator.
mi@0: 
mi@0:     Parameters
mi@0:     ----------
mi@0:     F : np.array((N,M))
mi@0:         Features matrix of N beats x M features.
mi@0:     M : int
mi@0:         Median filter size for the audio features (in beats).
mi@0:     Mg : int
mi@0:         Gaussian kernel size (in beats).
mi@0:     L : int
mi@0:         Median filter size for the adaptive threshold
mi@0: 
mi@0:     Return
mi@0:     ------
mi@0:     bound_idx : np.array
mi@0:         Array containing the indices of the boundaries.
mi@0:     """
mi@0:     # Filter
mitian@1:     F = SegUtil.median_filter(F, M=M)
mi@0: 
mi@0:     # Self Similarity Matrix
mitian@1:     S = SegUtil.compute_ssm(F)
mi@0: 
mi@0:     # Compute gaussian kernel
mitian@1:     G = SegUtil.compute_gaussian_krnl(Mg)
mi@0: 
mi@0:     # Compute the novelty curve
mitian@1:     nc = SegUtil.compute_nc(S, G)
mi@0: 
mi@0:     # Find peaks in the novelty curve
mitian@1:     return SegUtil.pick_peaks(nc, L=L, plot=plot)