{"id":13704814,"url":"https://github.com/bochinski/iou-tracker","last_synced_at":"2025-05-05T12:32:22.582Z","repository":{"id":201880429,"uuid":"98308472","full_name":"bochinski/iou-tracker","owner":"bochinski","description":"Python implementation of the IOU Tracker","archived":false,"fork":false,"pushed_at":"2020-02-18T13:03:22.000Z","size":41,"stargazers_count":688,"open_issues_count":20,"forks_count":176,"subscribers_count":31,"default_branch":"master","last_synced_at":"2024-05-22T08:33:34.419Z","etag":null,"topics":["demo-script","detrac","evaluation","iou-tracker","mot","python","tracker","tracking-by-detection","ua-detrac"],"latest_commit_sha":null,"homepage":"http://www.nue.tu-berlin.de","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/bochinski.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null}},"created_at":"2017-07-25T13:21:28.000Z","updated_at":"2024-04-15T04:41:12.000Z","dependencies_parsed_at":null,"dependency_job_id":"bde66dd6-48c2-4579-9344-d2b50d4be152","html_url":"https://github.com/bochinski/iou-tracker","commit_stats":null,"previous_names":["bochinski/iou-tracker"],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/bochinski%2Fiou-tracker","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/bochinski%2Fiou-tracker/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/bochinski%2Fiou-tracker/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/bochinski%2Fiou-tracker/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/bochinski","download_url":"https://codeload.github.com/bochinski/iou-tracker/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":224448696,"owners_count":17313102,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["demo-script","detrac","evaluation","iou-tracker","mot","python","tracker","tracking-by-detection","ua-detrac"],"created_at":"2024-08-02T22:00:17.594Z","updated_at":"2024-11-13T12:30:59.276Z","avatar_url":"https://github.com/bochinski.png","language":"Python","funding_links":[],"categories":["算法论文","Python"],"sub_categories":["**2017**"],"readme":"## IOU Tracker\nPython implementation of the IOU/V-IOU Tracker described in the AVSS 2017/2018 papers:\n\n[High-Speed Tracking-by-Detection Without Using Image Information](http://elvera.nue.tu-berlin.de/files/1517Bochinski2017.pdf)\n\n[Extending IOU Based Multi-Object Tracking by Visual Information](http://elvera.nue.tu-berlin.de/files/1547Bochinski2018.pdf)\n\nThis project is released under the MIT License (details in LICENSE file).\nIf you think our work is useful in your research, please consider citing:\n\n```\n@INPROCEEDINGS{1517Bochinski2017,\n\tAUTHOR = {Erik Bochinski and Volker Eiselein and Thomas Sikora},\n\tTITLE = {High-Speed Tracking-by-Detection Without Using Image Information},\n\tBOOKTITLE = {International Workshop on Traffic and Street Surveillance for Safety and Security at IEEE AVSS 2017},\n\tYEAR = {2017},\n\tMONTH = aug,\n\tADDRESS = {Lecce, Italy},\n\tURL = {http://elvera.nue.tu-berlin.de/files/1517Bochinski2017.pdf},\n\t}\n\n@INPROCEEDINGS{1547Bochinski2018,\n\tAUTHOR = {Erik Bochinski and Tobias Senst and Thomas Sikora},\n\tTITLE = {Extending IOU Based Multi-Object Tracking by Visual Information},\n\tBOOKTITLE = {IEEE International Conference on Advanced Video and Signals-based Surveillance},\n\tYEAR = {2018},\n\tMONTH = nov,\n\tPAGES = {441--446},\n\tADDRESS = {Auckland, New Zealand},\n\tURL = {http://elvera.nue.tu-berlin.de/files/1547Bochinski2018.pdf}\n}\n```\n### Table Of Contents\n- [Install](#Install)\n- [Demo](#Demo)\n- [DETRAC](#DETRAC)\n- [Motchallenge](#Motchallenge)\n  * [MOT16](#MOT16)\n  * [MOT17](#MOT17)\n  * [CVPR19](#CVPR19)\n- [Visdrone-MOT](#Visdrone-MOT)\n- [Contact](#Contact)\n\n**Update** (Jan 2020):\n* added V-IOU Code\n* updated README.md\n\n**Update** (December 2018):\n* added V-IOU results of our new paper [Extending IOU Based Multi-Object Tracking by Visual Information](http://elvera.nue.tu-berlin.de/files/1547Bochinski2018.pdf)\n* Mask R-CNN detections for UA-DETRAC added\n* CompACT parameters improved\n\n### Install\nThe repository now contains the code for both the IOU tracker and the V-IOU tracker.\nThe IOU Tracker only depends on numpy while V-IOU also requires OpenCV-Contrib and some other dependencies.\nIt is recommended to use a virtual environment to run the code:\n```\nvirtualenv -p python3 env\nsource env/bin/activate\npip install numpy lapsolver tqdm opencv-contrib-python\n```\nThis should get you started with a basic installation to run most the scripts in this repository.\n\n#### KCF/KCF2\nTwo different implementations of the KCF visual tracker can be used. One is supplied by OpenCV-Contrib and is denoted as *KCF*. This one should work out of the box.\nThe second implementation is denoted as *KCF2*. This one is needed to reproduce the reported results. In order to use this you need to install [KCFcpp-py-wrapper](https://github.com/uoip/KCFcpp-py-wrapper),\nwhich is a Cython based wrapper of the original KCF code for python.\nIt is recommended to build and install OpenCV from scratch instead of using the PyPI package in order to have all the necessary headers and libraries available.\n\nWhy? Because this implementation seems to work better and much faster than the one provided by OpenCV-Contrib. \n\nIf you do not install this module the tracker will automatically fall back on the OpenCV implementation (i.e. *KCF*) and a warning is printed. \nYou can get rid of this warning by either by installing the other KCF module as described above or explicitly request *KCF* instead of *KCF2* in the scripts. \nNote that the tracking performance will be affected by this.\n\n\n### Demo\nSeveral demo scripts are included to reproduce the reported results on the [UA-DETRAC](http://detrac-db.rit.albany.edu/)\n, [MOT](https://motchallenge.net/) 16/17/19 and [VisDrone](http://www.aiskyeye.com/)  benchmarks.\n\nBasic demo script (not dataset specific, can be used for other applications):\n```\n$ ./demo.py -h\nusage: demo.py [-h] [-v VISUAL] [-hr KEEP_UPPER_HEIGHT_RATIO] [-f FRAMES_PATH]\n               -d DETECTION_PATH -o OUTPUT_PATH [-sl SIGMA_L] [-sh SIGMA_H]\n               [-si SIGMA_IOU] [-tm T_MIN] [-ttl TTL] [-nms NMS] [-fmt FORMAT]\n\nIOU/V-IOU Tracker demo script\n\noptional arguments:\n  -h, --help            show this help message and exit\n  -v VISUAL, --visual VISUAL\n                        visual tracker for V-IOU. Currently supported are\n                        [BOOSTING, MIL, KCF, KCF2, TLD, MEDIANFLOW, GOTURN,\n                        NONE] see README.md for furthert details\n  -hr KEEP_UPPER_HEIGHT_RATIO, --keep_upper_height_ratio KEEP_UPPER_HEIGHT_RATIO\n                        Ratio of height of the object to track to the total\n                        height of the object for visual tracking. e.g. upper\n                        30%\n  -f FRAMES_PATH, --frames_path FRAMES_PATH\n                        sequence frames with format\n                        '/path/to/frames/frame_{:04d}.jpg' where '{:04d}' will\n                        be replaced with the frame id. (zero_padded to 4\n                        digits, use {:05d} for 5 etc.)\n  -d DETECTION_PATH, --detection_path DETECTION_PATH\n                        full path to CSV file containing the detections\n  -o OUTPUT_PATH, --output_path OUTPUT_PATH\n                        output path to store the tracking results (MOT\n                        challenge/Visdrone devkit compatible format)\n  -sl SIGMA_L, --sigma_l SIGMA_L\n                        low detection threshold\n  -sh SIGMA_H, --sigma_h SIGMA_H\n                        high detection threshold\n  -si SIGMA_IOU, --sigma_iou SIGMA_IOU\n                        intersection-over-union threshold\n  -tm T_MIN, --t_min T_MIN\n                        minimum track length\n  -ttl TTL, --ttl TTL   time to live parameter for v-iou\n  -nms NMS, --nms NMS   nms for loading multi-class detections\n  -fmt FORMAT, --format FORMAT\n                        format of the detections [motchallenge, visdrone]\n```\n\nExample for the IOU tracker on the MOT17-04 sequence (detections can be downloaded [here](https://motchallenge.net/data/MOT17/)):\n```\n./demo.py -d ../mot17/train/MOT17-04-SDP/det/det.txt -o res/iou-tracker/MOT17-04-SDP.txt\n```\n\nExample for the V-IOU tracker on the uav0000137_00458_v Visdrone sequence:\n```\ndemo.py -f '/path/to/VisDrone2018-MOT-val/sequences/uav0000137_00458_v/{:07d}.jpg' -d /path/to/VisDrone2018-MOT-val/detections/uav0000137_00458_v.txt -o results/VisDrone2018-MOT-val/uav0000137_00458_v.txt -v MEDIANFLOW -sl 0.9 -sh 0.98 -si 0.1 -tm 23 --ttl 8 --nms 0.6 -fmt visdrone\n```\n\n### DETRAC\nTo reproduce the reported results, download and extract the [DETRAC-toolkit](http://detrac-db.rit.albany.edu/download)\nand the detections you want to evaluate. Download links for the EB and Mask R-CNN detections are provided below.\nClone this repository into \"DETRAC-MOT-toolkit/trackers/\".\nFollow the instructions to configure the toolkit for tracking evaluation and set the tracker name in \"DETRAC_experiment.m\":\n\n```\ntracker.trackerName = 'iou-tracker';\n```\n\nand run the script.\nYou can switch between IOU and V-IOU and select the different parameters for different detectors in run_tracker.m \n\nNote that you still need a working python environment with numpy for IOU and all other dependencies for V-IOU installed.\nYou should obtain something like the following results for the 'DETRAC-Train' set:\n\n##### DETRAC-Train Results\nIOU Tracker:\n\n| Detector   | PR-Rcll | PR-Prcn | PR-FAR | PR-MT | PR-PT  | PR-ML | PR-FP   | PR-FN   | PR-IDs| PR-FM | PR-MOTA | PR-MOTP | PR-MOTAL |\n| --------   | ------- | ------- | ------ | ----- | ------ | ----- | ------- | ------- | ----- | ----- | ------- | ------- | -------- |\n| EB         |37.86    |44.73    |0.10    |32.34  |12.88   |20.93  |7958.82  |163739.85|4129.40|4221.89|35.77    |40.81    |36.48     |\n| R-CNN      |27.86    |52.90    |0.11    |19.53  |17.03   |18.56  |9047.95  |157521.18|4842.18|4969.57|25.46    |44.39    |26.29     |\n| CompACT    |25.20    |49.69    |0.10    |18.50  |14.11   |19.06  |8053.54  |153026.99|2021.84|2302.83|23.46    |42.96    |23.81     |\n| ACF        |27.39    |52.68    |0.14    |20.24  |15.66   |19.40  |11553.49 |161293.27|1845.49|2101.44|25.07    |44.71    |25.39     |\n| Mask R-CNN |43.21    |47.26    |0.60    |37.22  |11.46   |24.24  |50096.88 |171714.09|1021.94|929.53 |34.36    |45.43    |34.54     |\n\nV-IOU Tracker:\n\n| Detector   | PR-Rcll | PR-Prcn | PR-FAR | PR-MT | PR-PT  | PR-ML | PR-FP   | PR-FN   | PR-IDs| PR-FM | PR-MOTA | PR-MOTP | PR-MOTAL |\n| --------   | ------- | ------- | ------ | ----- | ------ | ----- | ------- | ------- | ----- | ----- | ------- | ------- | -------- |\n| CompACT    |26.84    |49.57    |0.10    |19.63  |14.67   |17.39  |8750.71  |143532.90|244.98 |444.20 |25.29    |41.58    |25.33     |\n| Mask R-CNN |42.80    |47.50    |0.60    |38.32  |8.36    |26.24  |50294.76 |174052.00|448.16 |293.69 |34.02    |46.87    |34.10     |\n\n\n##### DETRAC-Test (Overall) Results\nThe reference results are taken from the [UA-DETRAC results](http://detrac-db.rit.albany.edu/TraRet) site. Only the best tracker / detector\ncombination is displayed for each reference method.\n\n| Tracker       | Detector    | PR-MOTA | PR-MOTP     | PR-MT     | PR-ML     | PR-IDs   | PR-FM    | PR-FP      | PR-FN      | Speed          |\n| ------------- | ----------- | ------- | ----------- | --------- | --------- | -------- | -------- | ---------- | ---------- | -------------- |\n|CEM            | CompACT     | 5.1\\%     |35.2\\%     |3.0\\%      |35.3\\%     |267.9     |352.3     |**12341.2** |260390.4    |4.62 fps        |\n|CMOT           | CompACT     | 12.6\\%    |36.1\\%     |16.1\\%     |18.6\\%     |285.3     |1516.8    |57885.9     |167110.8    |3.79 fps        |\n|GOG            | CompACT     | 14.2\\%    |37.0\\%     |13.9\\%     |19.9\\%     |3334.6    |3172.4    |32092.9     |180183.8    |390 fps         |\n|DCT            | R-CNN       | 11.7\\%    |38.0\\%     |10.1\\%     |22.8\\%     |758.7     |742.9     |336561.2    |210855.6    |0.71 fps        |\n|H\u003csup\u003e2\u003c/sup\u003eT | CompACT     | 12.4\\%    |35.7\\%     |14.8\\%     |19.4\\%     |852.2     |1117.2    |51765.7     |173899.8    |3.02 fps        |\n|IHTLS          | CompACT     | 11.1\\%    |36.8\\%     |13.8\\%     |19.9\\%     |953.6     |3556.9    |53922.3     |180422.3    |19.79 fps       |\n|**IOU**        | R-CNN    |16.0\\%        |**38.3\\%** |13.8\\%     |20.7\\%     |5029.4    |5795.7    |22535.1     |193041.9    |100,840 fps     |\n|**IOU**        | EB       |19.4\\%        |28.9\\%     |17.7\\%     |**18.4\\%** |2311.3    |2445.9    |14796.5\t  |171806.8   |6,902 fps       |\n|**IOU**        | CompACT     | 16.1\\%    |37.0\\%     |14.8\\%     |19.7\\%     |2308.1    |3250.4    |24349.4     |176752.8    |**327,660 fps** |\n|**IOU**        | Mask R-CNN  | **30.7\\%**|37.0\\%     |30.3\\%     |21.5\\%     |668.0     |733.6     |17370.3     |179505.9    |14,956 fps      |\n|**V-IOU**      | CompACT     | 17.7\\%    |36.4\\%     |17.4\\%     |18.8\\%     |363.8     |1123.5    |26413.3     |**166571.7**|1117.90fps      |\n|**V-IOU**      | Mask R-CNN  | **30.7\\%**|37.0\\%     |**32.0\\%** |22.6\\%     |**162.6** |**286.2** |18046.2     |179191.2    |359.18 fps      |\n\n##### EB detections\nThe public detections of [EB](http://zyb.im/research/EB/) are not available on the\nDETRAC training set and miss some low scoring detections. The EB detections we used for the tables above and our\npublication are available here:\n\n* [EB Train](https://tubcloud.tu-berlin.de/s/EtC6cFEYsAU0gFQ/download)\n* [EB Test](https://tubcloud.tu-berlin.de/s/oKM3dYhJbMFl1dY/download)\n\n##### Mask R-CNN detections\nThese detections are generated using a recent Mask R-CNN implementation trained on COCO.\nOnly bounding boxes for COCOs *car*, *bus* and *truck* classes are included.\nNote that the detector is called \"frcnn\" (use `options.detectorSet = {'frcnn'};` in *initialize_environment.m*).\n* [Mask R-CNN Train](https://tubcloud.tu-berlin.de/s/MnGRGdH98WY9xQr/download)\n* [Mask R-CNN Test](https://tubcloud.tu-berlin.de/s/EztsFgm5AL8Jwtt/download)\n\n## Motchallenge\n\n### MOT16\nTo reproduce the reported [MOT16 results](https://motchallenge.net/results/MOT16/) of the paper, use the mot16.py script:\n\n```\n$ ./mot16.py -h\nusage: mot16.py [-h] -m SEQMAP -o RES_DIR -b BENCHMARK_DIR [-sl SIGMA_L]\n                [-sh SIGMA_H] [-si SIGMA_IOU] [-tm T_MIN]\n\nIOU Tracker MOT demo script. Default parameters are set to reproduce the\nresults using the SDP detections.\n\noptional arguments:\n  -h, --help            show this help message and exit\n  -m SEQMAP, --seqmap SEQMAP\n                        full path to the seqmap file to evaluate\n  -o RES_DIR, --res_dir RES_DIR\n                        path to the results directory\n  -b BENCHMARK_DIR, --benchmark_dir BENCHMARK_DIR\n                        path to the sequence directory\n  -sl SIGMA_L, --sigma_l SIGMA_L\n                        low detection threshold\n  -sh SIGMA_H, --sigma_h SIGMA_H\n                        high detection threshold\n  -si SIGMA_IOU, --sigma_iou SIGMA_IOU\n                        intersection-over-union threshold\n  -tm T_MIN, --t_min T_MIN\n                        minimum track length\n```\n\nExamples (you will probably need to adapt the paths):\n```\n# SDP:\n./mot16.py -m ../motchallenge/seqmaps/sdp-train.txt -o ../motchallenge/res/MOT16/iou-tracker -b ../data/mot17/train\n\n# FRCNN:\n./mot16.py -m ../motchallenge/seqmaps/frcnn-train.txt -o ../motchallenge/res/MOT16/iou-tracker -b ../data/mot17/train -sl 0 -sh 0.9 -si 0.3 -tm 5\n```\n\nThe seqmap files can be found under \"seqmaps\" and need to be copied to the respective directory of the\nmotchallenge devkit.\nYou should obtain something like the following results for the train set:\n\n##### MOT16 Train Results\n| Detector | IDF1 | IDP | IDR | Rcll | Prcn | FAR | GT  | MT  | PT  | ML  | FP  | FN  | IDs | FM  | MOTA | MOTP | MOTAL |\n| -------- | ---- | --- | --- | ---- | ---- | --- | --- | --- | --- | --- | --- | --- | --- | --- | ---- | ---- | ----- |\n|SDP       |24.7  |46.2 |16.9 |65.0  |97.6  |0.34 |546  |178  |232  |136  |1796 |39348|1198 |1453 |62.3  |83.4  |63.4   |\n|FRCNN     |21.0  |46.5 |13.6 |51.8  |97.2  |0.31 |546  |109  |261  |176  |1674 |54082|716  |810  |49.7  |88.2  |50.3   |\n\n##### MOT16 Test Results\n| Detector | Rcll | Prcn | FAR | GT  | MT  | PT  | ML  | FP  | FN  | IDs | FM  | MOTA | MOTP |\n| -------- | ---- | ---- | --- | --- | --- | --- | --- | --- | --- | --- | --- | ---- | ---- |\n|SDP       |61.5  |95.2  |0.96 |759  |179  |330  |250  |5702 |70278|2167 |3028 |57.1  |77.1  |\n|FRCNN     |50.9  |92.4  |1.29 |759  |113  |381  |265  |7639 |89535| 2284|2310 |45.4  |77.5  |\n\n Please note that this evaluation already includes the new ground truth of the MOT17 release.\n\n\n### MOT17\nThe IOU Tracker was evaluated on the MOT17 benchmark as well. To determine the best parameters for each detector, an\nexhaustive search of the parameter space was performed similar to the one of the MOT16 evaluation reported in the paper.\nThe best configuration for the training sequences is:\n\n| Detector | ![sigma_l](http://latex.codecogs.com/gif.latex?\\sigma_l)| ![sigma_h](http://latex.codecogs.com/gif.latex?\\sigma_h) | ![sigma_iou](http://latex.codecogs.com/gif.latex?\\sigma_{IOU}) | ![t_min](http://latex.codecogs.com/gif.latex?t_{min})      |\n| -------- | ----------------- | ----------------- | ------------------- | -------------------- |\n|DPM       | -0.5              | 0.5               | 0.4                 | 4                    |\n|FRCNN     | 0.0               | 0.9               | 0.3                 | 3                    |\n|SPD       | 0.4               | 0.5               | 0.2                 | 2                    |\n\nTo generate the MOT17 results listed at [MOT17 results](https://motchallenge.net/results/MOT17/), use the mot17.py script.\nNote that the parameters from above are hard-coded in the script for your convenience.\n```\nusage: mot17.py [-h] -m SEQMAP -o RES_DIR -b BENCHMARK_DIR\n\nIOU Tracker MOT17 demo script. The best parameters for each detector are\nhardcoded.\n\noptional arguments:\n  -h, --help            show this help message and exit\n  -m SEQMAP, --seqmap SEQMAP\n                        full path to the seqmap file to evaluate\n  -o RES_DIR, --res_dir RES_DIR\n                        path to the results directory\n  -b BENCHMARK_DIR, --benchmark_dir BENCHMARK_DIR\n                        path to the sequence directory\n```\n\nExamples (you will probably need to adapt the paths):\n```\n./mot17.py -m ../motchallenge/seqmaps/c10-train.txt -o ../motchallenge/res/MOT17/iou-tracker -b ../data/mot17/train\n./mot17.py -m ../motchallenge/seqmaps/c10-test.txt -o ../motchallenge/res/MOT17/iou-tracker -b ../data/mot17/test\n```\n\n##### MOT17 Train Results\n| Detector | IDF1 | IDP | IDR | Rcll | Prcn | FAR | GT  | MT  | PT  | ML  | FP  | FN  | IDs | FM  | MOTA | MOTP | MOTAL |\n| -------- | ---- | --- | --- | ---- | ---- | --- | --- | --- | --- | --- | --- | --- | --- | --- | ---- | ---- | ----- |\n|DPM       |14.3  |39.3 |8.7  | 35.8 | 88.1 | 1.02|546  |45   |195  |306  |5420 |72140|719  | 844 |  30.3|  77.1|  30.9 |\n|FRCNN     |21.6  |47.5 |14.0 | 52.1 | 97.0 | 0.34|546  |111  |268  |167  |1804 |53774|857  | 876 |  49.7|  88.1|  50.5 |\n|SDP       |24.4  |44.5 |16.8 | 66.8 | 96.8 | 0.47|546  |197  |240  |109  |2509 |37280|2058 | 2065|  62.7|  83.2|  64.6 |\n|All       |9.9   |21.5 |6.4  | 51.6 | 94.7 | 0.61|1638 |353  |703  |582  |97331|63194|3634 | 3785|  47.6|  83.4|  48.7 |\n\n##### MOT17 Test Results\n| MOTA | MOTP |\tFAF\t| MT    | ML    | FP    | FN     | ID Sw. | Frag |\n| ---- | ---- | --- | ----- | ----- | ----- | ------ | ------ | ---- |\n| 45.5 |76.9  |1.1\t|15.7\\% |40.5\\% |19,993\t|281,643 | 5,988  |7,404 |\n\n\n### CVPR19\nTo reproduce the results on the CVPR19 dataset you can use the cvpr19.sh bash script. \n\nEdit the first lines according to your setup:\n```\n# set these variables according to your setup\nseq_dir=/path/to/cvpr19/train # base directory of the split (cvpr19/train, cvpr19/test etc.)\nresults_dir=results/cvpr19    # output directory, will be created if not existing\n```\nThen, run (The used parameters for the *demo.py* script will be displayed for your convenience):\n```\n./cvpr19.sh\n```\n\nNote that this requires to have *KCF2* tracker installed.\n\nNote that only the upper 30% of the detections are used for visual tracking since the bottom part is often occluded.\nDue to time constraints only *KCF2* was evaluated for the CVPR19 challenge participation, \n*MEDIANFLOW* might yield better results like for the [Visdrone-MOT](#Visdrone-MOT) experiments but additional parameter tuning is required.\n\n##### CVPR19 Train Results\n| Detector | IDF1 | IDP | IDR | Rcll | Prcn | FAR | GT  | MT  | PT  | ML  | FP  | FN  | IDs | FM  | MOTA | MOTP | MOTAL |\n| -------- | ---- | --- | --- | ---- | ---- | --- | --- | --- | --- | --- | --- | --- | --- | --- | ---- | ---- | ----- |\n| FRCNN    | 53.9 | 71.6| 43.2| 59.9 | 99.3 | 0.63| 2274| 670 |1187 | 417 |55844|97170| 3272|4792 | 59.2 | 87.5 | 59.5  |\n\n### Visdrone-MOT\nTo reproduce the results on the Visdrone MOT dataset you can use the visdrone-mot.sh bash script.\nEdit the first lines according to your setup:\n```\n# set these variables according to your setup\nvisdrone_dir=/path/to/VisDrone2018-MOT-val  # base directory of the split (VisDrone2018-MOT-val, VisDrone2018-MOT-train etc.)\nresults_dir=results/VisDrone2018-MOT-val    # output directory, will be created if not existing\nvis_tracker=MEDIANFLOW                      # [MEDIANFLOW, KCF2, NONE] parameter set as used in the paper\n```\nThen, run (The used parameters for the *demo.py* script will be displayed for your convenience):\n```\n./visdrone-mot.sh\n```\n\n##### Visdrone-MOT Val Results\nFor the *VisDrone2018-MOT-val* split you should get the following results:\n\n| Visual Tracker | IDF1 | IDP | IDR | Rcll | Prcn | FAR | GT  | MT  | PT  | ML  | FP  | FN  | IDs | FM  | MOTA | MOTP | MOTAL |\n| -------------- | ---- | --- | --- | ---- | ---- | --- | --- | --- | --- | --- | --- | --- | --- | --- | ---- | ---- | ----- |\n| None           | 40.9 | 68.5| 29.2| 34.6 | 81.3 | 0.60| 476 | 102 | 76  | 297 | 5736|46979| 177 | 435 | 26.4 | 78.1 | 26.6  |\n| KCF            | 45.3 | 75.3| 32.4| 35.2 | 81.8 | 0.59| 476 | 105 | 64  | 305 | 5605|46578| 76  | 385 | 27.3 | 77.9 | 27.4  |\n| Medianflow     | 45.6 | 75.9| 32.6| 35.3 | 82.2 | 0.58| 476 | 107 | 63  | 304 | 5494|46466| 65  | 378 | 27.6 | 77.8 | 27.7  |\n\n\n## Contact\nIf you have any questions or encounter problems regarding the method/code feel free to contact me\nat bochinski@nue.tu-berlin.de\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fbochinski%2Fiou-tracker","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fbochinski%2Fiou-tracker","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fbochinski%2Fiou-tracker/lists"}