Mercurial > hg > nsaunier > traffic-intelligence

comparison scripts/dltrack.py @ 1230:c582b272108f

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
(minor) work in progress
author Nicolas Saunier <nicolas.saunier@polymtl.ca>
date Mon, 21 Aug 2023 15:49:32 -0400
parents 759d76d6d20c
children 6487ef10c0e0
comparison
equal deleted inserted replaced
1229:759d76d6d20c 1230:c582b272108f
10 # detect model 10 # detect model
11 # tracker model 11 # tracker model
12 parser.add_argument('--display', dest = 'display', help = 'show the results (careful with long videos, risk of running out of memory)', action = 'store_true') 12 parser.add_argument('--display', dest = 'display', help = 'show the results (careful with long videos, risk of running out of memory)', action = 'store_true')
13 args = parser.parse_args() 13 args = parser.parse_args()
14 14
15 # required functionality?
16 # # filename of the video to process (can be images, eg image%04d.png)
17 # video-filename = laurier.avi
18 # # filename of the database where results are saved
19 # database-filename = laurier.sqlite
20 # # filename of the homography matrix
21 # homography-filename = laurier-homography.txt
22 # # filename of the camera intrinsic matrix
23 # intrinsic-camera-filename = intrinsic-camera.txt
24 # # -0.11759321 0.0148536 0.00030756 -0.00020578 -0.00091816
25 # distortion-coefficients = -0.11759321
26 # distortion-coefficients = 0.0148536
27 # distortion-coefficients = 0.00030756
28 # distortion-coefficients = -0.00020578
29 # distortion-coefficients = -0.00091816
30 # # undistorted image multiplication
31 # undistorted-size-multiplication = 1.31
32 # # Interpolation method for remapping image when correcting for distortion: 0 for INTER_NEAREST - a nearest-neighbor interpolation; 1 for INTER_LINEAR - a bilinear interpolation (used by default); 2 for INTER_CUBIC - a bicubic interpolation over 4x4 pixel neighborhood; 3 for INTER_LANCZOS4
33 # interpolation-method = 1
34 # # filename of the mask image (where features are detected)
35 # mask-filename = none
36 # # undistort the video for feature tracking
37 # undistort = false
38 # # load features from database
39 # load-features = false
40 # # display trajectories on the video
41 # display = false
42 # # original video frame rate (number of frames/s)
43 # video-fps = 29.97
44 # # number of digits of precision for all measurements derived from video
45 # # measurement-precision = 3
46 # # first frame to process
47 # frame1 = 0
48 # # number of frame to process: 0 means processing all frames
49 # nframes = 0
50
51 # TODO add option to refine position with mask for vehicles
52
15 # Load a model 53 # Load a model
16 model = YOLO('/home/nicolas/Research/Data/classification-models/yolov8x.pt') # seg yolov8x-seg.pt 54 model = YOLO('/home/nicolas/Research/Data/classification-models/yolov8x.pt') # seg yolov8x-seg.pt
17 # seg could be used on cropped image... if can be loaded and kept in memory 55 # seg could be used on cropped image... if can be loaded and kept in memory
18 # model = YOLO('/home/nicolas/Research/Data/classification-models/yolo_nas_l.pt ') # AttributeError: 'YoloNAS_L' object has no attribute 'get' 56 # model = YOLO('/home/nicolas/Research/Data/classification-models/yolo_nas_l.pt ') # AttributeError: 'YoloNAS_L' object has no attribute 'get'
19 57
21 if args.display: 59 if args.display:
22 results = model.track(source=args.videoFilename, tracker="/home/nicolas/Research/Data/classification-models/bytetrack.yaml", classes=list(cocoTypeNames.keys()), show=True) # , save_txt=True 60 results = model.track(source=args.videoFilename, tracker="/home/nicolas/Research/Data/classification-models/bytetrack.yaml", classes=list(cocoTypeNames.keys()), show=True) # , save_txt=True
23 else: 61 else:
24 results = model.track(source=args.videoFilename, tracker="/home/nicolas/Research/Data/classification-models/bytetrack.yaml", classes=list(cocoTypeNames.keys()), stream=True) 62 results = model.track(source=args.videoFilename, tracker="/home/nicolas/Research/Data/classification-models/bytetrack.yaml", classes=list(cocoTypeNames.keys()), stream=True)
25 for result in results: 63 for result in results:
64 print(len(result.boxes))
26 for box in result.boxes: 65 for box in result.boxes:
27 print(box.xyxy) 66 print(box.xyxy)