Mercurial > hg > nsaunier > traffic-intelligence

comparison scripts/dltrack.py @ 1234:dd969637381e

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
work on tracker interface
author Nicolas Saunier <nicolas.saunier@polymtl.ca>
date Thu, 14 Sep 2023 16:18:36 -0400
parents d5695e0b59d9
children 855abc69fa99
comparison
equal deleted inserted replaced
1233:d5695e0b59d9 1234:dd969637381e
6 import cv2 6 import cv2
7 7
8 from trafficintelligence import cvutils, moving, storage, utils 8 from trafficintelligence import cvutils, moving, storage, utils
9 9
10 parser = argparse.ArgumentParser(description='The program tracks objects following the ultralytics yolo executable.')#, epilog = 'Either the configuration filename or the other parameters (at least video and database filenames) need to be provided.') 10 parser = argparse.ArgumentParser(description='The program tracks objects following the ultralytics yolo executable.')#, epilog = 'Either the configuration filename or the other parameters (at least video and database filenames) need to be provided.')
11 parser.add_argument('-i', dest = 'videoFilename', help = 'name of the video file (overrides the configuration file)') 11 parser.add_argument('-i', dest = 'videoFilename', help = 'name of the video file', required = True)
12 # detect model 12 parser.add_argument('-d', dest = 'databaseFilename', help = 'name of the Sqlite database file', required = True)
13 # tracker model 13 parser.add_argument('-m', dest = 'detectorFilename', help = 'name of the detection model file', required = True)
14 parser.add_argument('-t', dest = 'trackerFilename', help = 'name of the tracker file', required = True)
14 parser.add_argument('--display', dest = 'display', help = 'show the results (careful with long videos, risk of running out of memory)', action = 'store_true') 15 parser.add_argument('--display', dest = 'display', help = 'show the results (careful with long videos, risk of running out of memory)', action = 'store_true')
15 #parser.add_argument('-f', dest = 'firstFrameNum', help = 'show the results (careful with long videos, risk of running out of memory)', action = 'store_true') 16 parser.add_argument('-f', dest = 'firstFrameNum', help = 'number of first frame number to process', type = int, default = 0)
17 parser.add_argument('-l', dest = 'lastFrameNum', help = 'number of last frame number to process', type = int, default = float('Inf'))
16 args = parser.parse_args() 18 args = parser.parse_args()
17 19
18 # required functionality? 20 # required functionality?
19 # # filename of the video to process (can be images, eg image%04d.png) 21 # # filename of the video to process (can be images, eg image%04d.png)
20 # video-filename = laurier.avi 22 # video-filename = laurier.avi
57 59
58 60
59 # check if one can go to specific frame https://docs.ultralytics.com/modes/track/#persisting-tracks-loop 61 # check if one can go to specific frame https://docs.ultralytics.com/modes/track/#persisting-tracks-loop
60 62
61 # Load a model 63 # Load a model
62 model = YOLO('/home/nicolas/Research/Data/classification-models/yolov8x.pt') # seg yolov8x-seg.pt 64 model = YOLO('/home/nicolas/Research/Data/classification-models/yolov8x.pt', ) # seg yolov8x-seg.pt
63 # seg could be used on cropped image... if can be loaded and kept in memory 65 # seg could be used on cropped image... if can be loaded and kept in memory
64 # model = YOLO('/home/nicolas/Research/Data/classification-models/yolo_nas_l.pt ') # AttributeError: 'YoloNAS_L' object has no attribute 'get' 66 # model = YOLO('/home/nicolas/Research/Data/classification-models/yolo_nas_l.pt ') # AttributeError: 'YoloNAS_L' object has no attribute 'get'
65 67
66 # Track with the model 68 # Track with the model
67 #results = model.track(source=args.videoFilename, tracker="/home/nicolas/Research/Data/classification-models/bytetrack.yaml", classes=list(moving.cocoTypeNames.keys()), show=True) # , save_txt=True 69 #results = model.track(source=args.videoFilename, tracker="/home/nicolas/Research/Data/classification-models/bytetrack.yaml", classes=list(moving.cocoTypeNames.keys()), show=True) # , save_txt=True
68 if args.display: 70 if args.display:
69 windowName = 'frame' 71 windowName = 'frame'
70 cv2.namedWindow(windowName, cv2.WINDOW_NORMAL) 72 cv2.namedWindow(windowName, cv2.WINDOW_NORMAL)
71 73
72 results = model.track(source=args.videoFilename, tracker="/home/nicolas/Research/Data/classification-models/bytetrack.yaml", classes=list(moving.cocoTypeNames.keys()), stream=True) 74 capture = cv2.VideoCapture(args.videoFilename)
75 #results = model.track(source=args.videoFilename, tracker="/home/nicolas/Research/Data/classification-models/bytetrack.yaml", classes=list(moving.cocoTypeNames.keys()), stream=True)
73 objects = [] 76 objects = []
74 currentObjects = {} 77 currentObjects = {}
75 featureNum = 0 78 featureNum = 0
79
80 frameNum = args.firstFrameNum
81 capture.set(cv2.CAP_PROP_POS_FRAMES, frameNum)
82 lastFrameNum = args.lastFrameNum
83
84 success, frame = capture.read()
85 results = model.track(frame, tracker="/home/nicolas/Research/Data/classification-models/bytetrack.yaml", classes=list(moving.cocoTypeNames.keys()), persist=True)
76 # create object with user type and list of 3 features (bottom ones and middle) + projection 86 # create object with user type and list of 3 features (bottom ones and middle) + projection
77 for frameNum, result in enumerate(results): 87 while capture.isOpened() and success and frameNum <= lastFrameNum:
88 #for frameNum, result in enumerate(results):
89 result = results[0]
78 print(frameNum, len(result.boxes)) 90 print(frameNum, len(result.boxes))
79 for box in result.boxes: 91 for box in result.boxes:
80 #print(box.cls, box.id, box.xyxy) 92 #print(box.cls, box.id, box.xyxy)
81 if box.id is not None: # None are objects with low confidence 93 if box.id is not None: # None are objects with low confidence
82 num = int(box.id) 94 num = int(box.id)
100 if args.display: 112 if args.display:
101 cvutils.cvImshow(windowName, result.plot()) # original image in orig_img 113 cvutils.cvImshow(windowName, result.plot()) # original image in orig_img
102 key = cv2.waitKey() 114 key = cv2.waitKey()
103 if cvutils.quitKey(key): 115 if cvutils.quitKey(key):
104 break 116 break
117 frameNum += 1
118 success, frame = capture.read()
119 results = model.track(frame, persist=True)
105 120
106 # interpolate before saving 121 # interpolate and generate velocity before saving
107 for num, obj in currentObjects.items(): 122 for num, obj in currentObjects.items():
108 obj.setUserType(utils.mostCommon(obj.userTypes)) 123 obj.setUserType(utils.mostCommon(obj.userTypes))
109 obj.features[0].timeInterval = copy(obj.getTimeInterval()) 124 obj.features[0].timeInterval = copy(obj.getTimeInterval())
110 obj.features[1].timeInterval = copy(obj.getTimeInterval()) 125 obj.features[1].timeInterval = copy(obj.getTimeInterval())
111 if obj.length() != len(obj.features[0].tmpPositions): # interpolate 126 if obj.length() != len(obj.features[0].tmpPositions): # interpolate
113 obj.features[1].positions = moving.Trajectory.fromPointDict(obj.features[1].tmpPositions) 128 obj.features[1].positions = moving.Trajectory.fromPointDict(obj.features[1].tmpPositions)
114 else: 129 else:
115 obj.features[0].positions = moving.Trajectory.fromPointList(list(obj.features[0].tmpPositions.values())) 130 obj.features[0].positions = moving.Trajectory.fromPointList(list(obj.features[0].tmpPositions.values()))
116 obj.features[1].positions = moving.Trajectory.fromPointList(list(obj.features[1].tmpPositions.values())) 131 obj.features[1].positions = moving.Trajectory.fromPointList(list(obj.features[1].tmpPositions.values()))
117 132
118 storage.saveTrajectoriesToSqlite('test.sqlite', list(currentObjects.values()), 'object') 133 storage.saveTrajectoriesToSqlite(args.databaseFilename, list(currentObjects.values()), 'object')
119 134
120 # todo save bbox and mask to study localization / representation 135 # todo save bbox and mask to study localization / representation
121 # apply quality checks deviation and acceleration bounds? 136 # apply quality checks deviation and acceleration bounds?