pose2sim/Pose2Sim/filtering.py

#!/usr/bin/env python
# -*- coding: utf-8 -*-


'''
    ###########################################################################
    ## FILTER 3D COORDINATES                                                 ##
    ###########################################################################
    
    Filter trc 3D coordinates.

    Available filters: Butterworth, Butterworth on speed, Gaussian, LOESS, Median
    Set your parameters in Config.toml
        
    INPUTS: 
    - a trc file
    - filtering parameters in Config.toml
    
    OUTPUT: 
    - a filtered trc file
    
'''


## INIT
import os
import fnmatch
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import logging

from scipy import signal
from scipy.ndimage import gaussian_filter1d
from statsmodels.nonparametric.smoothers_lowess import lowess
from filterpy.kalman import KalmanFilter, rts_smoother
from filterpy.common import Q_discrete_white_noise

from Pose2Sim.common import plotWindow

## AUTHORSHIP INFORMATION
__author__ = "David Pagnon"
__copyright__ = "Copyright 2021, Pose2Sim"
__credits__ = ["David Pagnon"]
__license__ = "BSD 3-Clause License"
__version__ = '0.4'
__maintainer__ = "David Pagnon"
__email__ = "contact@david-pagnon.com"
__status__ = "Development"


## FUNCTIONS
def kalman_filter(coords, frame_rate, measurement_noise, process_noise, nb_dimensions=3, nb_derivatives=3, smooth=True):
    '''
    Filters coordinates with a Kalman filter or a Kalman smoother
    
    INPUTS:
    - coords: array of shape (nframes, ndims)
    - frame_rate: integer
    - measurement_noise: integer
    - process_noise: integer
    - nb_dimensions: integer, number of dimensions (3 if 3D coordinates)
    - nb_derivatives: integer, number of derivatives (3 if constant acceleration model)
    - smooth: boolean. True if souble pass (recommended), False if single pass (if real-time)
    
    OUTPUTS:
    - kpt_coords_filt: filtered coords
    '''

    # Variables
    dim_x = nb_dimensions * nb_derivatives # 9 state variables 
    dt = 1/frame_rate
    
    # Filter definition
    f = KalmanFilter(dim_x=dim_x, dim_z=nb_dimensions)

    # States: initial position, velocity, accel, in 3D
    def derivate_array(arr, dt=1):
        return np.diff(arr, axis=0)/dt
    def repeat(func, arg_func, nb_reps):
        for i in range(nb_reps):
            arg_func = func(arg_func)
        return arg_func
    x_init = []
    for n_der in range(nb_derivatives):
        x_init += [repeat(derivate_array, coords, n_der)[0]] # pose*3D, vel*3D, accel*3D
    f.x = np.array(x_init).reshape(nb_dimensions,nb_derivatives).T.flatten() # pose, vel, accel *3D
    
    # State transition matrix
    F_per_coord = np.zeros((int(dim_x/nb_dimensions), int(dim_x/nb_dimensions)))
    for i in range(nb_derivatives):
        for j in range(min(i+1, nb_derivatives)):
            F_per_coord[j,i] = dt**(i-j) / np.math.factorial(i - j)
    f.F = np.kron(np.eye(nb_dimensions),F_per_coord) 
    # F_per_coord= [[1, dt, dt**2/2], 
                 # [ 0, 1,  dt     ],
                 # [ 0, 0,  1      ]])

    # No control input
    f.B = None 

    # Measurement matrix (only positions)
    H = np.zeros((nb_dimensions, dim_x)) 
    for i in range(min(nb_dimensions,dim_x)):
        H[i, int(i*(dim_x/nb_dimensions))] = 1
    f.H = H
    # H = [[1., 0., 0., 0., 0., 0., 0., 0., 0.],
        # [0., 0., 0., 1., 0., 0., 0., 0., 0.],
        # [0., 0., 0., 0., 0., 0., 1., 0., 0.]]

    # Covariance matrix
    f.P *= measurement_noise 

    # Measurement noise
    f.R = np.diag([measurement_noise**2]*nb_dimensions) 

    # Process noise
    f.Q = Q_discrete_white_noise(nb_derivatives, dt=dt, var=process_noise**2, block_size=nb_dimensions) 

    # Run filter: predict and update for each frame
    mu, cov, _, _ = f.batch_filter(coords) # equivalent to below
    # mu = []
    # for kpt_coord_frame in coords:
        # f.predict()
        # f.update(kpt_coord_frame)
        # mu.append(f.x.copy())
    ind_of_position = [int(d*(dim_x/nb_dimensions)) for d in range(nb_dimensions)]
    coords_filt = np.array(mu)[:,ind_of_position]

    # RTS smoother
    if smooth == True:
        mu2, P, C, _ = f.rts_smoother(mu, cov)
        coords_filt = np.array(mu2)[:,ind_of_position]

    return coords_filt


def kalman_filter_1d(config, col):
    '''
    1D Kalman filter
    Deals with nans
    
    INPUT:
    - col: Pandas dataframe column
    - trustratio: int, ratio process_noise/measurement_noise
    - framerate: int
    - smooth: boolean, True if double pass (recommended), False if single pass (if real-time)

    OUTPUT:
    - col_filtered: Filtered pandas dataframe column
    '''

    trustratio = int(config.get('filtering').get('kalman').get('trust_ratio'))
    smooth = int(config.get('filtering').get('kalman').get('smooth'))
    framerate = config.get('project').get('frame_rate')
    measurement_noise = 20
    process_noise = measurement_noise * trustratio

    # split into sequences of not nans
    col_filtered = col.copy()
    mask = np.isnan(col_filtered)  | col_filtered.eq(0)
    falsemask_indices = np.where(~mask)[0]
    gaps = np.where(np.diff(falsemask_indices) > 1)[0] + 1 
    idx_sequences = np.split(falsemask_indices, gaps)
    if idx_sequences[0].size > 0:
        idx_sequences_to_filter = [seq for seq in idx_sequences]
    
        # Filter each of the selected sequences
        for seq_f in idx_sequences_to_filter:
            col_filtered[seq_f] = kalman_filter(col_filtered[seq_f], framerate, measurement_noise, process_noise, nb_dimensions=1, nb_derivatives=3, smooth=smooth).flatten()

    return col_filtered


def butterworth_filter_1d(config, col):
    '''
    1D Zero-phase Butterworth filter (dual pass)
    Deals with nans

    INPUT:
    - col: numpy array
    - order: int
    - cutoff: int
    - framerate: int

    OUTPUT:
    - col_filtered: Filtered pandas dataframe column
    '''

    type = 'low' #config.get('filtering').get('butterworth').get('type')
    order = int(config.get('filtering').get('butterworth').get('order'))
    cutoff = int(config.get('filtering').get('butterworth').get('cut_off_frequency'))    
    framerate = config.get('project').get('frame_rate')

    b, a = signal.butter(order/2, cutoff/(framerate/2), type, analog = False) 
    padlen = 3 * max(len(a), len(b))
    
    # split into sequences of not nans
    col_filtered = col.copy()
    mask = np.isnan(col_filtered)  | col_filtered.eq(0)
    falsemask_indices = np.where(~mask)[0]
    gaps = np.where(np.diff(falsemask_indices) > 1)[0] + 1 
    idx_sequences = np.split(falsemask_indices, gaps)
    if idx_sequences[0].size > 0:
        idx_sequences_to_filter = [seq for seq in idx_sequences if len(seq) > padlen]
    
        # Filter each of the selected sequences
        for seq_f in idx_sequences_to_filter:
            col_filtered[seq_f] = signal.filtfilt(b, a, col_filtered[seq_f])
    
    return col_filtered
    

def butterworth_on_speed_filter_1d(config, col):
    '''
    1D zero-phase Butterworth filter (dual pass) on derivative

    INPUT:
    - col: Pandas dataframe column
    - frame rate, order, cut-off frequency, type (from Config.toml)

    OUTPUT:
    - col_filtered: Filtered pandas dataframe column
    '''

    type = 'low' # config.get('filtering').get('butterworth_on_speed').get('type')
    order = int(config.get('filtering').get('butterworth_on_speed').get('order'))
    cutoff = int(config.get('filtering').get('butterworth_on_speed').get('cut_off_frequency'))
    framerate = config.get('project').get('frame_rate')

    b, a = signal.butter(order/2, cutoff/(framerate/2), type, analog = False)
    padlen = 3 * max(len(a), len(b))
    
    # derivative
    col_filtered = col.copy()
    col_filtered_diff = col_filtered.diff()   # derivative
    col_filtered_diff = col_filtered_diff.fillna(col_filtered_diff.iloc[1]/2) # set first value correctly instead of nan
    
    # split into sequences of not nans
    mask = np.isnan(col_filtered_diff)  | col_filtered_diff.eq(0)
    falsemask_indices = np.where(~mask)[0]
    gaps = np.where(np.diff(falsemask_indices) > 1)[0] + 1 
    idx_sequences = np.split(falsemask_indices, gaps)
    if idx_sequences[0].size > 0:
        idx_sequences_to_filter = [seq for seq in idx_sequences if len(seq) > padlen]
    
        # Filter each of the selected sequences
        for seq_f in idx_sequences_to_filter:
            col_filtered_diff[seq_f] = signal.filtfilt(b, a, col_filtered_diff[seq_f])
    col_filtered = col_filtered_diff.cumsum() + col.iloc[0] # integrate filtered derivative
    
    return col_filtered


def gaussian_filter_1d(config, col):
    '''
    1D Gaussian filter

    INPUT:
    - col: Pandas dataframe column
    - gaussian_filter_sigma_kernel: kernel size from Config.toml

    OUTPUT:
    - col_filtered: Filtered pandas dataframe column
    '''

    gaussian_filter_sigma_kernel = int(config.get('filtering').get('gaussian').get('sigma_kernel'))

    col_filtered = gaussian_filter1d(col, gaussian_filter_sigma_kernel)

    return col_filtered
    

def loess_filter_1d(config, col):
    '''
    1D LOWESS filter (Locally Weighted Scatterplot Smoothing)

    INPUT:
    - col: Pandas dataframe column
    - loess_filter_nb_values: window used for smoothing from Config.toml
    frac = loess_filter_nb_values * frames_number

    OUTPUT:
    - col_filtered: Filtered pandas dataframe column
    '''

    kernel = config.get('filtering').get('LOESS').get('nb_values_used')

    col_filtered = col.copy()
    mask = np.isnan(col_filtered) 
    falsemask_indices = np.where(~mask)[0]
    gaps = np.where(np.diff(falsemask_indices) > 1)[0] + 1 
    idx_sequences = np.split(falsemask_indices, gaps)
    if idx_sequences[0].size > 0:
        idx_sequences_to_filter = [seq for seq in idx_sequences if len(seq) > kernel]
    
        # Filter each of the selected sequences
        for seq_f in idx_sequences_to_filter:
            col_filtered[seq_f] = lowess(col_filtered[seq_f], seq_f, is_sorted=True, frac=kernel/len(seq_f), it=0)[:,1]

    return col_filtered
    

def median_filter_1d(config, col):
    '''
    1D median filter

    INPUT:
    - col: Pandas dataframe column
    - median_filter_kernel_size: kernel size from Config.toml
    
    OUTPUT:
    - col_filtered: Filtered pandas dataframe column
    '''
    
    median_filter_kernel_size = config.get('filtering').get('median').get('kernel_size')
    
    col_filtered = signal.medfilt(col, kernel_size=median_filter_kernel_size)

    return col_filtered


def display_figures_fun(Q_unfilt, Q_filt, time_col, keypoints_names):
    '''
    Displays filtered and unfiltered data for comparison

    INPUTS:
    - Q_unfilt: pandas dataframe of unfiltered 3D coordinates
    - Q_filt: pandas dataframe of filtered 3D coordinates
    - time_col: pandas column
    - keypoints_names: list of strings

    OUTPUT:
    - matplotlib window with tabbed figures for each keypoint
    '''

    pw = plotWindow()
    for id, keypoint in enumerate(keypoints_names):
        f = plt.figure()
        
        axX = plt.subplot(311)
        plt.plot(time_col.to_numpy(), Q_unfilt.iloc[:,id*3].to_numpy(), label='unfiltered')
        plt.plot(time_col.to_numpy(), Q_filt.iloc[:,id*3].to_numpy(), label='filtered')
        plt.setp(axX.get_xticklabels(), visible=False)
        axX.set_ylabel(keypoint+' X')
        plt.legend()

        axY = plt.subplot(312)
        plt.plot(time_col.to_numpy(), Q_unfilt.iloc[:,id*3+1].to_numpy(), label='unfiltered')
        plt.plot(time_col.to_numpy(), Q_filt.iloc[:,id*3+1].to_numpy(), label='filtered')
        plt.setp(axY.get_xticklabels(), visible=False)
        axY.set_ylabel(keypoint+' Y')
        plt.legend()

        axZ = plt.subplot(313)
        plt.plot(time_col.to_numpy(), Q_unfilt.iloc[:,id*3+2].to_numpy(), label='unfiltered')
        plt.plot(time_col.to_numpy(), Q_filt.iloc[:,id*3+2].to_numpy(), label='filtered')
        axZ.set_ylabel(keypoint+' Z')
        axZ.set_xlabel('Time')
        plt.legend()

        pw.addPlot(keypoint, f)
    
    pw.show()


def filter1d(col, config, filter_type):
    '''
    Choose filter type and filter column

    INPUT:
    - col: Pandas dataframe column
    - filter_type: filter type from Config.toml
    
    OUTPUT:
    - col_filtered: Filtered pandas dataframe column
    '''

    # Choose filter
    filter_mapping = {
        'kalman': kalman_filter_1d,
        'butterworth': butterworth_filter_1d, 
        'butterworth_on_speed': butterworth_on_speed_filter_1d, 
        'gaussian': gaussian_filter_1d, 
        'LOESS': loess_filter_1d, 
        'median': median_filter_1d
        }
    filter_fun = filter_mapping[filter_type]
    
    # Filter column
    col_filtered = filter_fun(config, col)

    return col_filtered


def recap_filter3d(config, trc_path):
    '''
    Print a log message giving filtering parameters. Also stored in User/logs.txt.

    OUTPUT:
    - Message in console
    '''

    # Read Config
    filter_type = config.get('filtering').get('type')
    kalman_filter_trustratio = int(config.get('filtering').get('kalman').get('trust_ratio'))
    kalman_filter_smooth = int(config.get('filtering').get('kalman').get('smooth'))
    kalman_filter_smooth_str = 'smoother' if kalman_filter_smooth else 'filter'
    butterworth_filter_type = 'low' # config.get('filtering').get('butterworth').get('type')
    butterworth_filter_order = int(config.get('filtering').get('butterworth').get('order'))
    butterworth_filter_cutoff = int(config.get('filtering').get('butterworth').get('cut_off_frequency'))
    butter_speed_filter_type = 'low' # config.get('filtering').get('butterworth_on_speed').get('type')
    butter_speed_filter_order = int(config.get('filtering').get('butterworth_on_speed').get('order'))
    butter_speed_filter_cutoff = int(config.get('filtering').get('butterworth_on_speed').get('cut_off_frequency'))
    gaussian_filter_sigma_kernel = int(config.get('filtering').get('gaussian').get('sigma_kernel'))
    loess_filter_nb_values = config.get('filtering').get('LOESS').get('nb_values_used')
    median_filter_kernel_size = config.get('filtering').get('median').get('kernel_size')

    # Recap
    filter_mapping_recap = {
        'kalman': f'--> Filter type: Kalman {kalman_filter_smooth_str}. Measurements trusted {kalman_filter_trustratio} times as much as previous data, assuming a constant acceleration process.', 
        'butterworth': f'--> Filter type: Butterworth {butterworth_filter_type}-pass. Order {butterworth_filter_order}, Cut-off frequency {butterworth_filter_cutoff} Hz.', 
        'butterworth_on_speed': f'--> Filter type: Butterworth on speed {butter_speed_filter_type}-pass. Order {butter_speed_filter_order}, Cut-off frequency {butter_speed_filter_cutoff} Hz.', 
        'gaussian': f'--> Filter type: Gaussian. Standard deviation kernel: {gaussian_filter_sigma_kernel}', 
        'LOESS': f'--> Filter type: LOESS. Number of values used: {loess_filter_nb_values}', 
        'median': f'--> Filter type: Median. Kernel size: {median_filter_kernel_size}'
    }
    logging.info(filter_mapping_recap[filter_type])
    logging.info(f'Filtered 3D coordinates are stored at {trc_path}.')


def filter_all(config):
    '''
    Filter the 3D coordinates of the trc file.
    Displays filtered coordinates for checking.

    INPUTS:
    - a trc file
    - filtration parameters from Config.toml

    OUTPUT:
    - a filtered trc file
    '''

    # Read config
    project_dir = config.get('project').get('project_dir')
    try:
        pose_tracked_dir = os.path.join(project_dir, 'pose-associated')
        os.listdir(pose_tracked_dir)
        pose_dir = pose_tracked_dir
    except:
        pose_dir = os.path.join(project_dir, 'pose')
    frame_range = config.get('project').get('frame_range')
    seq_name = os.path.basename(os.path.realpath(project_dir))
    pose3d_dir = os.path.realpath(os.path.join(project_dir, 'pose-3d'))
    display_figures = config.get('filtering').get('display_figures')
    filter_type = config.get('filtering').get('type')

    # Frames range
    pose_listdirs_names = next(os.walk(pose_dir))[1]
    json_dirs_names = [k for k in pose_listdirs_names if 'json' in k]
    json_files_names = [fnmatch.filter(os.listdir(os.path.join(pose_dir, js_dir)), '*.json') for js_dir in json_dirs_names]
    f_range = [[0,min([len(j) for j in json_files_names])] if frame_range==[] else frame_range][0]
    
    # Trc paths
    trc_f_in = f'{seq_name}_{f_range[0]}-{f_range[1]}.trc'
    trc_f_in_LSTM = f'{seq_name}_{f_range[0]}-{f_range[1]}_LSTM.trc'
    trc_f_out = f'{seq_name}_filt_{filter_type}_{f_range[0]}-{f_range[1]}.trc'
    trc_f_out_LSTM = f'{seq_name}_filt_{filter_type}_{f_range[0]}-{f_range[1]}_LSTM.trc'
    if os.path.exists(os.path.join(pose3d_dir, trc_f_in_LSTM)):
        trc_path_in = os.path.join(pose3d_dir, trc_f_in_LSTM)
        trc_path_out = os.path.join(pose3d_dir, trc_f_out_LSTM)
    elif os.path.exists(os.path.join(pose3d_dir, trc_f_in)):
        trc_path_in = os.path.join(pose3d_dir, trc_f_in)
        trc_path_out = os.path.join(pose3d_dir, trc_f_out)
    else:
        raise FileNotFoundError("Not found .trc file.")
    
    # Read trc header
    with open(trc_path_in, 'r') as trc_file:
        header = [next(trc_file) for line in range(5)]

    # Read trc coordinates values
    trc_df = pd.read_csv(trc_path_in, sep="\t", skiprows=4)
    frames_col, time_col = trc_df.iloc[:,0], trc_df.iloc[:,1]
    Q_coord = trc_df.drop(trc_df.columns[[0, 1]], axis=1)

    # Filter coordinates
    Q_filt = Q_coord.apply(filter1d, axis=0, args = [config, filter_type])

    # Display figures
    if display_figures:
        # Retrieve keypoints
        keypoints_names = pd.read_csv(trc_path_in, sep="\t", skiprows=3, nrows=0).columns[2::3].to_numpy()
        display_figures_fun(Q_coord, Q_filt, time_col, keypoints_names)

    # Reconstruct trc file with filtered coordinates
    with open(trc_path_out, 'w') as trc_o:
        [trc_o.write(line) for line in header]
        Q_filt.insert(0, 'Frame#', frames_col)
        Q_filt.insert(1, 'Time', time_col)
        Q_filt.to_csv(trc_o, sep='\t', index=False, header=None, lineterminator='\n')

    # Recap
    recap_filter3d(config, trc_path_out)