sort跟踪

 

sort_wp

"""
    SORT: A Simple, Online and Realtime Tracker
    Copyright (C) 2016 Alex Bewley alex@dynamicdetection.com

    This program is free software: you can redistribute it and/or modify
    it under the terms of the GNU General Public License as published by
    the Free Software Foundation, either version 3 of the License, or
    (at your option) any later version.

    This program is distributed in the hope that it will be useful,
    but WITHOUT ANY WARRANTY; without even the implied warranty of
    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
    GNU General Public License for more details.

    You should have received a copy of the GNU General Public License
    along with this program.  If not, see <http://www.gnu.org/licenses/>.
"""
from __future__ import print_function

import os
import numpy as np
import matplotlib.pyplot as plt
import matplotlib.patches as patches
from skimage import io
from sklearn.utils.linear_assignment_ import linear_assignment
import glob
import time
import argparse
from filterpy.kalman import KalmanFilter

def iou(a, b):
    """
    Computes IUO between two bboxes in the form [x1,y1,x2,y2]
    """
    xx1 = np.maximum(a[0], b[0])
    yy1 = np.maximum(a[1], b[1])
    xx2 = np.minimum(a[2], b[2])
    yy2 = np.minimum(a[3], b[3])
    w = np.maximum(0., xx2 - xx1)
    h = np.maximum(0., yy2 - yy1)
    inter = w * h
    aarea = (a[2] - a[0]) * (a[3] - a[1])
    barea = (b[2] - b[0]) * (b[3] - b[1])
    o = inter / (aarea + barea - inter)
    return(o)

def nms(dets, thresh):
    x1 = dets[:, 0]
    y1 = dets[:, 1]
    x2 = dets[:, 2]
    y2 = dets[:, 3]
    scores = dets[:, 4]

    areas = (x2 - x1 + 1) * (y2 - y1 + 1)
    order = scores.argsort()[::-1]

    keep = []
    while order.size > 0:
        i = order[0]
        keep.append(i)
        xx1 = np.maximum(x1[i], x1[order[1:]])
        yy1 = np.maximum(y1[i], y1[order[1:]])
        xx2 = np.minimum(x2[i], x2[order[1:]])
        yy2 = np.minimum(y2[i], y2[order[1:]])

        w = np.maximum(0.0, xx2 - xx1 + 1)
        h = np.maximum(0.0, yy2 - yy1 + 1)
        inter = w * h
        ovr = inter / (areas[i] + areas[order[1:]] - inter)

        inds = np.where(ovr <= thresh)[0]
        order = order[inds + 1]

    return keep

def convert_bbox_to_z(bbox):
    """
    Takes a bounding box in the form [x1,y1,x2,y2] and returns z in the form
        [x,y,s,r] where x,y is the centre of the box and s is the scale/area and r is
        the aspect ratio
    """
    w = bbox[2] - bbox[0]
    h = bbox[3] - bbox[1]
    x = bbox[0] + w / 2.
    y = bbox[1] + h / 2.
    s = w * h    #scale is just area
    r = w / float(h)
    return np.array([x, y, s, r]).reshape((4, 1))

def convert_x_to_bbox(x,score=None):
    """
    Takes a bounding box in the centre form [x,y,s,r] and returns it in the form
        [x1,y1,x2,y2] where x1,y1 is the top left and x2,y2 is the bottom right
    """
    w = np.sqrt(x[2]*x[3])
    h = x[2] / w
    if(score == None):
        return np.array([x[0]-w/2.,x[1]-h/2.,x[0]+w/2.,x[1]+h/2.]).reshape((1,4))
    else:
        return np.array([x[0]-w/2.,x[1]-h/2.,x[0]+w/2.,x[1]+h/2.,score]).reshape((1,5))


class KalmanBoxTracker(object):
    """
    This class represents the internel state of individual tracked objects observed as bbox.
    """
    count = 0
    def __init__(self, bbox, cls_=0):
        """
        Initialises a tracker using initial bounding box.
        """
        #define constant velocity model
        self.kf = KalmanFilter(dim_x=7, dim_z=4)
        self.kf.F = np.array([[1,0,0,0,1,0,0],[0,1,0,0,0,1,0],[0,0,1,0,0,0,1],[0,0,0,1,0,0,0],  [0,0,0,0,1,0,0],[0,0,0,0,0,1,0],[0,0,0,0,0,0,1]])
        self.kf.H = np.array([[1,0,0,0,0,0,0],[0,1,0,0,0,0,0],[0,0,1,0,0,0,0],[0,0,0,1,0,0,0]])

        self.kf.R[2:,2:] *= 10.
        self.kf.P[4:,4:] *= 1000. #give high uncertainty to the unobservable initial velocities
        self.kf.P *= 10.
        self.kf.Q[-1,-1] *= 0.01
        self.kf.Q[4:,4:] *= 0.01

        self.kf.x[:4] = convert_bbox_to_z(bbox)
        self.time_since_update = 0
        self.id = KalmanBoxTracker.count
        KalmanBoxTracker.count += 1
        self.history = []
        self.hits = 0
        self.hit_streak = 0
        self.age = 0

        self.cls_ = cls_

    def update(self, bbox, cls_=0):
        """
        Updates the state vector with observed bbox.
        """
        self.time_since_update = 0
        self.history = []
        self.hits += 1
        self.hit_streak += 1
        self.kf.update(convert_bbox_to_z(bbox))
    self.cls_ = cls_

    def predict(self):
        """
        Advances the state vector and returns the predicted bounding box estimate.
        """
        if((self.kf.x[6] + self.kf.x[2])<=0):
            self.kf.x[6] *= 0.0
        self.kf.predict()
        self.age += 1
        if(self.time_since_update>0):
            self.hit_streak = 0
        self.time_since_update += 1
        self.history.append(convert_x_to_bbox(self.kf.x))
        return self.history[-1]

    def get_state(self):
        """
        Returns the current bounding box estimate.
        """
        return convert_x_to_bbox(self.kf.x, 0.8)

def associate_detections_to_trackers(detections, trackers, iou_threshold = 0.7):
    """
    Assigns detections to tracked object (both represented as bounding boxes)
    Returns 3 lists of matches, unmatched_detections and unmatched_trackers
    """
    if(len(trackers)==0):
        return np.empty((0,2),dtype=int), np.arange(len(detections)), np.empty((0,5),dtype=int)
    iou_matrix = np.zeros((len(detections),len(trackers)),dtype=np.float32)

    for d,det in enumerate(detections):
        for t,trk in enumerate(trackers):
            iou_matrix[d,t] = iou(det,trk)
    matched_indices = linear_assignment(-iou_matrix)

    unmatched_detections = []
    for d,det in enumerate(detections):
        if(d not in matched_indices[:,0]):
            unmatched_detections.append(d)
  
    unmatched_trackers = []
    for t,trk in enumerate(trackers):
        if(t not in matched_indices[:,1]):
            unmatched_trackers.append(t)

    #filter out matched with low IOU
    matches = []
    for m in matched_indices:
        if(iou_matrix[m[0],m[1]]<iou_threshold):
            unmatched_detections.append(m[0])
            unmatched_trackers.append(m[1])
        else:
            matches.append(m.reshape(1,2))
    if(len(matches)==0):
        matches = np.empty((0,2),dtype=int)
    else:
        matches = np.concatenate(matches,axis=0)

    print(matches)
    print(unmatched_detections)
    print(unmatched_trackers)

    return matches, np.array(unmatched_detections), np.array(unmatched_trackers)



class Sort(object):
  
    def __init__(self, max_age=1, min_hits=3):
        """
        Sets key parameters for SORT
        """
        self.max_age = max_age
        self.min_hits = min_hits
        self.trackers = []
        self.frame_count = 0

    def update(self, dets):
        """
        Params:
            dets - a numpy array of detections in the format [[x1,y1,x2,y2,score],[x1,y1,x2,y2,score],...]
            Requires: this method must be called once for each frame even with empty detections.
            Returns the a similar array, where the last column is the object ID.

        NOTE: The number of objects returned may differ from the number of detections provided.
        """
        boxes, cls = dets[:, :5], dets[:, 5]
        dets = boxes
        print("------------------")
        print(dets[:, :4])
        print("------------------")
        self.frame_count += 1
        #get predicted locations from existing trackers.
        trks = np.zeros((len(self.trackers), 5))
        # print(trks)
        to_del = []
        ret = []
        n_box = []
        # print(self.trackers)
        for t,trk in enumerate(trks):
            # print(t)
            # print(trk)
            pos = self.trackers[t].predict()[0]
            trk[:] = [pos[0], pos[1], pos[2], pos[3], 0]
            if(np.any(np.isnan(pos))):
                to_del.append(t)
        trks = np.ma.compress_rows(np.ma.masked_invalid(trks))
        """
    keep = nms(trks, 0.3)   
        for t in range(len(trks)):
        if t not in to_del and t not in keep:
        to_del.append(t)
    """
    for t in reversed(to_del):
            self.trackers.pop(t)
        matched, unmatched_dets, unmatched_trks = associate_detections_to_trackers(dets,trks)

        #update matched trackers with assigned detections
        for t,trk in enumerate(self.trackers):
            if(t not in unmatched_trks):
                d = matched[np.where(matched[:,1]==t)[0],0]
                trk.update(dets[d,:][0], cls[d])

        #create and initialise new trackers for unmatched detections
        for i in unmatched_dets:
            trk = KalmanBoxTracker(dets[i,:], cls[i]) 
            self.trackers.append(trk)
        i = len(self.trackers)
        print('trackers cnt:', i)
    for trk in reversed(self.trackers):
        # for trk in self.trackers:
            d = trk.get_state()[0]
        # if(trk.time_since_update < 1):
        #     n_box.append([int(d[0]), int(d[1]), int(d[2]-d[0]), int(d[3]-d[1])])
            print("update since", trk.time_since_update)
        # print("frame cnt:", self.frame_count)
        print("hit streak:", trk.hit_streak)
        if((trk.time_since_update < 1) and (trk.hit_streak == 0 or trk.hit_streak >= self.min_hits or self.frame_count <= self.min_hits)):
            # ret.append(np.concatenate((d, [trk.id+1])).reshape(1,-1)) # +1 as MOT benchmark requires positive
        ret.append(np.concatenate((d, [trk.cls_, trk.id+1])).reshape(1,-1)) # +1 as MOT benchmark requires positive
            n_box.append([int(d[0]), int(d[1]), int(d[2]-d[0]), int(d[3]-d[1])])
        i -= 1
            #remove dead tracklet
            if(trk.time_since_update > self.max_age):
                self.trackers.pop(i)
        print("xxxxxxxxxxxxxxxxxxxxxxx")
        print(n_box)
        print("xxxxxxxxxxxxxxxxxxxxxxx")
    if(len(ret)>0):
            return np.concatenate(ret)
        return np.empty((0, 6))
    
# def parse_args():
#     """Parse input arguments."""
#     parser = argparse.ArgumentParser(description='SORT demo')
#     parser.add_argument('--display', dest='display', help='Display online tracker output (slow) [False]',action='store_true')
#     args = parser.parse_args()
#     return args

if __name__ == '__main__':
    # all train
    # sequences = ['PETS09-S2L1','TUD-Campus','TUD-Stadtmitte','ETH-Bahnhof','ETH-Sunnyday','ETH-Pedcross2','KITTI-13','KITTI-17','ADL-Rundle-6','ADL-Rundle-8','Venice-2']
    # IMGS_DIR = './236frames'
    DET_DIR = './MOT'
    # args = parse_args()
    # phase = 'train'
    total_time = 0.0
    total_frames = 0
    # colours = np.random.rand(32,3) #used only for display
  
    if not os.path.exists('output'):
        os.makedirs('output')
  
    mot_tracker = Sort() #create instance of the SORT tracker

    # imgs_fn_list = [os.path.join(IMGS_DIR, fn) for fn in sorted(os.listdir(IMGS_DIR))]
    dets_fn_list = [os.path.join(DET_DIR, fn) for fn in sorted(os.listdir(DET_DIR))]
    # print(dets_fn_list)
    
    fout = open("output/adas.txt", "w")
    for i in range(len(dets_fn_list)):
    # for i in range(210):
        total_frames += 1
        print("process #", total_frames)
        dets = np.loadtxt(dets_fn_list[i], delimiter=' ')
        # print(dets)
        # det_boxes = []
        if len(dets)>0:
            if dets.ndim == 1:
                dets = dets[np.newaxis, :]
            # dets = dets[:, :5]
        # det_boxes = dets
        dets[:, 2:4] += dets[:, 0:2] # convert to [x1,y1,w,h] to [x1,y1,x2,y2]
        else:
            dets = np.empty((0, 6), dtype=int)
        # print(dets.shape)
        start_time = time.time()
        trackers = mot_tracker.update(dets)
        cycle_time = time.time() - start_time
        total_time += cycle_time
    
    boxes = trackers[:, :6].tolist()
    # for k in range(len(det_boxes)):
    #     x1, y1, x2, y2, score, c = det_boxes[k, :]
    #     boxes.append([x1, y1, x2, y2, score, c])
    # boxes = np.asarray(boxes)
    # keep = nms(boxes, 0.2)
    # boxes = boxes[keep, :]
    for b in boxes:
        fout.write("%d %d %d %d %d %d\n"%(i+1, int(b[0]), int(b[1]), int(b[2]-b[0]), int(b[3]-b[1]), int(b[5])))
    """
        for d in trackers:
            fout.write('%d %d %d %d %d %d %d\n'%(i+1, int(d[4]), int(d[0]), int(d[1]), int(d[2]-d[0]), int(d[3]-d[1]), int(d[5])))
        """
    fout.close()

    print("Total Tracking took: %.3f s for %d frames or %.1f FPS"%(total_time, total_frames, total_frames / total_time))

 

输入的格式是x1,y1,w,h,prob,cls,也就是所谓的MOT格式