出现以下Error: image 1/2001 /home/fish/AIC21-MTMC/datasets/detection/images/test/S06/c043/img1/img000000.jpg: Traceback (most recent call last): File "detect2img.py", line 235, in detect() File "detect2img.py", line 95, in detect pred = model(img, augment=opt.augment)[0] File "/home/fish/anaconda3/envs/aic21/lib/python3.8/site-packages/torch/nn/modules/module.py", line 727, in _call_impl result = self.forward(*input, **kwargs) File "/home/fish/AIC21-MTMC/detector/yolov5/models/yolo.py", line 119, in forward return self.forward_once(x, profile) # single-scale inference, train File "/home/fish/AIC21-MTMC/detector/yolov5/models/yolo.py", line 135, in forward_once x = m(x) # run File "/home/fish/anaconda3/envs/aic21/lib/python3.8/site-packages/torch/nn/modules/module.py", line 727, in _call_impl result = self.forward(*input, **kwargs) File "/home/fish/AIC21-MTMC/detector/yolov5/models/yolo.py", line 54, in forward y[..., 2:4] = (y[..., 2:4] * 2) ** 2 * self.anchor_grid[i] # wh RuntimeError: The size of tensor a (160) must match the size of tensor b (48) at non-singleton dimension 3

想请问一下我应该如何处理？

I'm reproducing the run_all results and got some questions.

In the run_all.sh, the re-id feature extraction part is as below, python extract_image_feat.py "aic_reid1.yml" python extract_image_feat.py "aic_reid2.yml" python extract_image_feat.py "aic_reid3.yml"

Checked that the config files differ in the pretrained re-id model each uses, resnet101 or resnext101. So I wonder should I run all the three re-id feature extraction processes or just one of them?

Further noticed in reid/merge_reid_feat.py, in merge_feat(), ensemble_list = ['detect_reid1', 'detect_reid2', 'detect_reid3']. I guess this ensemble_list should be in accordance with the re-id models selected. If I only use reid1, should I modify the ensemble_list to contain only the 'detect_reid1'?

Thank you for your patience and regards!

As is stated in the README, a detect_provided folder which contains detections and Re-ID features should be put in the dataset path. However there is no link for downloading this. Would you please fix it and provide a link? Thank you very much and regards!

In method associate in tracker/MOTBaseline/src/post_processing/post_association.py, the external C++ program MinCostPerfMatch is called with argument "--max", which solves maximum cardinality matching. The computed cost (similarity) is not used in this step.

Should it instead be solving min cost perfect matching problem, which calls MinCostPerfMatch with argument --minweight?

Thanks.

I did not find any links to access the following resources: "If you just want reproduce our results, you can directly download detect_provided" "If you just want reproduce our results on all pipeline, you have to download:" """ detector/yolov5/yolov5x.pt reid/reid_model/resnet101_ibn_a_2.pth reid/reid_model/resnet101_ibn_a_3.pth reid/reid_model/resnext101_ibn_a_2.pth """ how can I get them?

When I run run_mcmt.sh, the following error occurs. Do you have any suggestions?

starting re_ranking
100%|█████████████████████████████████████████████████████████████████████████████████████| 78/78 [00:00<00:00, 6529.14it/s]
Using totally 0.06S to compute R
100%|█████████████████████████████████████████████████████████████████████████████████████| 78/78 [00:00<00:00, 7608.45it/s]
Using totally 0.10S to compute V-1
Using totally 0.12S to compute V-2
Using totally 0.12S to compute invIndex
100%|█████████████████████████████████████████████████████████████████████████████████████| 39/39 [00:00<00:00, 6944.80it/s]
[[nan nan nan ... nan nan nan]
 [nan nan nan ... nan nan nan]
 [nan nan nan ... nan nan nan]
 ...
 [nan nan nan ... nan nan nan]
 [nan nan nan ... nan nan nan]
 [nan nan nan ... nan nan nan]]
Using totally 0.15S to compute final_distance
[[nan nan nan ... nan nan nan]
 [nan nan nan ... nan nan nan]
 [nan nan nan ... nan nan nan]
 ...
 [nan nan nan ... nan nan nan]
 [nan nan nan ... nan nan nan]
 [nan nan nan ... nan nan nan]]
Traceback (most recent call last):
  File "sub_cluster.py", line 154, in <module>
    clu = get_labels(cfg,cid_tid_dict,cid_tids,score_thr=cfg.SCORE_THR)
  File "sub_cluster.py", line 106, in get_labels
    linkage='complete').fit_predict(1 - sim_matrix)
  File "/home/ubuntu/anaconda3/envs/torch/lib/python3.7/site-packages/sklearn/cluster/_agglomerative.py", line 1054, in fit_predict
    return super().fit_predict(X, y)
  File "/home/ubuntu/anaconda3/envs/torch/lib/python3.7/site-packages/sklearn/base.py", line 736, in fit_predict
    self.fit(X)
  File "/home/ubuntu/anaconda3/envs/torch/lib/python3.7/site-packages/sklearn/cluster/_agglomerative.py", line 917, in fit
    X = self._validate_data(X, ensure_min_samples=2, estimator=self)
  File "/home/ubuntu/anaconda3/envs/torch/lib/python3.7/site-packages/sklearn/base.py", line 566, in _validate_data
    X = check_array(X, **check_params)
  File "/home/ubuntu/anaconda3/envs/torch/lib/python3.7/site-packages/sklearn/utils/validation.py", line 800, in check_array
    _assert_all_finite(array, allow_nan=force_all_finite == "allow-nan")
  File "/home/ubuntu/anaconda3/envs/torch/lib/python3.7/site-packages/sklearn/utils/validation.py", line 116, in _assert_all_finite
    type_err, msg_dtype if msg_dtype is not None else X.dtype
ValueError: Input contains NaN, infinity or a value too large for dtype('float32').
Traceback (most recent call last):
  File "gen_res.py", line 40, in <module>
    map_tid = pickle.load(open('test_cluster.pkl', 'rb'))['cluster']
FileNotFoundError: [Errno 2] No such file or directory: 'test_cluster.pkl'

I ran your all pipeline on AIC, but the program crashed when fork eight GPU processes in re-id. I realized there're only two GPUs in my system. How should I configure the number of GPUs to be implemented? Thank you very much and regards!

When trying to run the entire pipeline, I receive the following error: Traceback (most recent call last): File "merge_reid_feat.py", line 57, in main() File "merge_reid_feat.py", line 53, in main merge_feat(cfg) File "merge_reid_feat.py", line 25, in merge_feat feat_mode_dic = pickle.load(open(feat_pkl_file, 'rb')) FileNotFoundError: [Errno 2] No such file or directory: '/home/----/aic21-mtmc/datasets/detect_reid1/c041/c041_dets_feat.pkl'

Any thoughts?

Congratulations and thanks for your great project!

When I run the visualization for mcmt (viz_mcmt.py), I got the errors as following:

start cam:41 [ERROR:0] global /tmp/pip-req-build-947ayiyu/opencv/modules/videoio/src/cap.cpp (160) open VIDEOIO(CV_IMAGES): raised OpenCV exception: OpenCV(4.5.2) /tmp/pip-req-build-947ayiyu/opencv/modules/videoio/src/cap_images.cpp:253: error: (-5:Bad argument) CAP_IMAGES: can't find starting number (in the name of file): /gpfs/gsfs1/home/tnguyen/coding/reid_all/AIC21-MTMC/datasets/AIC21_Track3_MTMC_Tracking/mcmt/S06/c041/vdo.avi in function 'icvExtractPattern'

Are there any steps that we need to run before this visualization ?

Thanks for your great project, and I was trying to reproduce the results. It said I can directly download the "detect_provided", but I couldn't find where to download it. I tried to run the run_all.sh, and thought that the folder might be generated after that. However I encountered too many errors, and I am still working on it.

Congratulations and thanks for your great project!

I am investigating your code and paper. I see that there are TransReId module in reid model. Did you run the experiment with TransReId feats instead of feats from resnet backbone ? If yes, is the result competitive ? If no, are continue working on that ?

How can I use your work to matching local tracklets across multiple cameras to generate global tracks for person tracking? I have tracklets sets from each camera from strongsort tracker, but I want to use them to generate global trajectories, I would appreciate your help.

Thanks for sharing the great project, I am trying to reproduce the results. When I run bash ./run_mcmt.sh, I met a error as follows: dell@DESKTOP-291GB2L MINGW64 /d/THICV/Project/AIC21-MTMC (main) $ bash ./run_mcmt.sh [Err]: invalid source root Done tracking c041 with aic_mcmt.yml tracking c042 with aic_mcmt.yml tracking c043 with aic_mcmt.yml tracking c044 with aic_mcmt.yml tracking c045 with aic_mcmt.yml tracking c046 with aic_mcmt.yml Traceback (most recent call last): File "D:\THICV\Project\AIC21-MTMC\tracker\MOTBaseline\src\fair_app.py", line 309, in Traceback (most recent call last): File "D:\THICV\Project\AIC21-MTMC\tracker\MOTBaseline\src\fair_app.py", line 309, in run( File "D:\THICV\Project\AIC21-MTMC\tracker\MOTBaseline\src\fair_app.py", line 175, in run Traceback (most recent call last): File "D:\THICV\Project\AIC21-MTMC\tracker\MOTBaseline\src\fair_app.py", line 309, in run( seq_info = gather_sequence_info(sequence_dir, detection_file, max_frame_idx) File "D:\THICV\Project\AIC21-MTMC\tracker\MOTBaseline\src\fair_app.py", line 175, in run File "D:\THICV\Project\AIC21-MTMC\tracker\MOTBaseline\src\fair_app.py", line 49, in gather_sequence_info Traceback (most recent call last): File "D:\THICV\Project\AIC21-MTMC\tracker\MOTBaseline\src\fair_app.py", line 309, in for f in os.listdir(image_dir)} FileNotFoundError: [WinError 3] ϵͳ▒Ҳ▒▒▒ָ▒▒▒▒·▒▒▒▒: 'D:/THICV/Porject/AIC21_MTMC/datasets/detection/images/test/S06/c046\img1' run( File "D:\THICV\Project\AIC21-MTMC\tracker\MOTBaseline\src\fair_app.py", line 175, in run seq_info = gather_sequence_info(sequence_dir, detection_file, max_frame_idx) File "D:\THICV\Project\AIC21-MTMC\tracker\MOTBaseline\src\fair_app.py", line 49, in gather_sequence_info seq_info = gather_sequence_info(sequence_dir, detection_file, max_frame_idx) File "D:\THICV\Project\AIC21-MTMC\tracker\MOTBaseline\src\fair_app.py", line 49, in gather_sequence_info run( File "D:\THICV\Project\AIC21-MTMC\tracker\MOTBaseline\src\fair_app.py", line 175, in run for f in os.listdir(image_dir)} FileNotFoundError: [WinError 3] ϵͳ▒Ҳ▒▒▒ָ▒▒▒▒·▒▒▒▒: 'D:/THICV/Porject/AIC21_MTMC/datasets/detection/images/test/S06/c043\img1' for f in os.listdir(image_dir)} FileNotFoundError: [WinError 3] ϵͳ▒Ҳ▒▒▒ָ▒▒▒▒·▒▒▒▒: 'D:/THICV/Porject/AIC21_MTMC/datasets/detection/images/test/S06/c042\img1' seq_info = gather_sequence_info(sequence_dir, detection_file, max_frame_idx) File "D:\THICV\Project\AIC21-MTMC\tracker\MOTBaseline\src\fair_app.py", line 49, in gather_sequence_info for f in os.listdir(image_dir)} FileNotFoundError: [WinError 3] ϵͳ▒Ҳ▒▒▒ָ▒▒▒▒·▒▒▒▒: 'D:/THICV/Porject/AIC21_MTMC/datasets/detection/images/test/S06/c044\img1' Traceback (most recent call last): File "D:\THICV\Project\AIC21-MTMC\tracker\MOTBaseline\src\fair_app.py", line 309, in run( File "D:\THICV\Project\AIC21-MTMC\tracker\MOTBaseline\src\fair_app.py", line 175, in run seq_info = gather_sequence_info(sequence_dir, detection_file, max_frame_idx) File "D:\THICV\Project\AIC21-MTMC\tracker\MOTBaseline\src\fair_app.py", line 49, in gather_sequence_info for f in os.listdir(image_dir)} FileNotFoundError: [WinError 3] ϵͳ▒Ҳ▒▒▒ָ▒▒▒▒·▒▒▒▒: 'D:/THICV/Porject/AIC21_MTMC/datasets/detection/images/test/S06/c041\img1' Traceback (most recent call last): File "D:\THICV\Project\AIC21-MTMC\tracker\MOTBaseline\src\fair_app.py", line 309, in run( File "D:\THICV\Project\AIC21-MTMC\tracker\MOTBaseline\src\fair_app.py", line 175, in run seq_info = gather_sequence_info(sequence_dir, detection_file, max_frame_idx) File "D:\THICV\Project\AIC21-MTMC\tracker\MOTBaseline\src\fair_app.py", line 49, in gather_sequence_info for f in os.listdir(image_dir)} FileNotFoundError: [WinError 3] ϵͳ▒Ҳ▒▒▒ָ▒▒▒▒·▒▒▒▒: 'D:/THICV/Porject/AIC21_MTMC/datasets/detection/images/test/S06/c045\img1' sys args ars: ['main.py', 'c044', 'pp', 'aic_mcmt.yml'] using post processing loading tracked file D:/THICV/Porject/AIC21_MTMC/datesets/detect_provided/c044/c044_mot.txt Traceback (most recent call last): File "D:\THICV\Project\AIC21-MTMC\tracker\MOTBaseline\src\post_processing\main.py", line 110, in eval_seq(sys.argv[1], pp=sys.argv[2], mcmt_cfg=cfg) File "D:\THICV\Project\AIC21-MTMC\tracker\MOTBaseline\src\post_processing\main.py", line 41, in eval_seq results = load_raw_mot(seq, mcmt_cfg) File "D:\THICV\Project\AIC21-MTMC\tracker\MOTBaseline\src\post_processing\main.py", line 17, in load_raw_mot mot_feat_dic = pickle.load(open(f'{mcmt_cfg.DATA_DIR}/{seq}/{seq}_mot_feat_raw.pkl', 'rb')) FileNotFoundError: [Errno 2] No such file or directory: 'D:/THICV/Porject/AIC21_MTMC/datesets/detect_provided/c044/c044_mot_feat_raw.pkl' sys args ars: ['main.py', 'c045', 'pp', 'aic_mcmt.yml'] using post processing loading tracked file D:/THICV/Porject/AIC21_MTMC/datesets/detect_provided/c045/c045_mot.txt Traceback (most recent call last): File "D:\THICV\Project\AIC21-MTMC\tracker\MOTBaseline\src\post_processing\main.py", line 110, in sys args ars: ['main.py', 'c041', 'pp', 'aic_mcmt.yml'] using post processing loading tracked file D:/THICV/Porject/AIC21_MTMC/datesets/detect_provided/c041/c041_mot.txt Traceback (most recent call last): File "D:\THICV\Project\AIC21-MTMC\tracker\MOTBaseline\src\post_processing\main.py", line 110, in eval_seq(sys.argv[1], pp=sys.argv[2], mcmt_cfg=cfg) File "D:\THICV\Project\AIC21-MTMC\tracker\MOTBaseline\src\post_processing\main.py", line 41, in eval_seq eval_seq(sys.argv[1], pp=sys.argv[2], mcmt_cfg=cfg) File "D:\THICV\Project\AIC21-MTMC\tracker\MOTBaseline\src\post_processing\main.py", line 41, in eval_seq results = load_raw_mot(seq, mcmt_cfg) File "D:\THICV\Project\AIC21-MTMC\tracker\MOTBaseline\src\post_processing\main.py", line 17, in load_raw_mot results = load_raw_mot(seq, mcmt_cfg) File "D:\THICV\Project\AIC21-MTMC\tracker\MOTBaseline\src\post_processing\main.py", line 17, in load_raw_mot mot_feat_dic = pickle.load(open(f'{mcmt_cfg.DATA_DIR}/{seq}/{seq}_mot_feat_raw.pkl', 'rb')) FileNotFoundError: [Errno 2] No such file or directory: 'D:/THICV/Porject/AIC21_MTMC/datesets/detect_provided/c045/c045_mot_feat_raw.pkl' mot_feat_dic = pickle.load(open(f'{mcmt_cfg.DATA_DIR}/{seq}/{seq}_mot_feat_raw.pkl', 'rb')) FileNotFoundError: [Errno 2] No such file or directory: 'D:/THICV/Porject/AIC21_MTMC/datesets/detect_provided/c041/c041_mot_feat_raw.pkl' sys args ars: ['main.py', 'c042', 'pp', 'aic_mcmt.yml'] using post processing loading tracked file D:/THICV/Porject/AIC21_MTMC/datesets/detect_provided/c042/c042_mot.txt Traceback (most recent call last): File "D:\THICV\Project\AIC21-MTMC\tracker\MOTBaseline\src\post_processing\main.py", line 110, in sys args ars: ['main.py', 'c046', 'pp', 'aic_mcmt.yml'] using post processing loading tracked file D:/THICV/Porject/AIC21_MTMC/datesets/detect_provided/c046/c046_mot.txt Traceback (most recent call last): File "D:\THICV\Project\AIC21-MTMC\tracker\MOTBaseline\src\post_processing\main.py", line 110, in eval_seq(sys.argv[1], pp=sys.argv[2], mcmt_cfg=cfg) File "D:\THICV\Project\AIC21-MTMC\tracker\MOTBaseline\src\post_processing\main.py", line 41, in eval_seq eval_seq(sys.argv[1], pp=sys.argv[2], mcmt_cfg=cfg) File "D:\THICV\Project\AIC21-MTMC\tracker\MOTBaseline\src\post_processing\main.py", line 41, in eval_seq results = load_raw_mot(seq, mcmt_cfg) File "D:\THICV\Project\AIC21-MTMC\tracker\MOTBaseline\src\post_processing\main.py", line 17, in load_raw_mot results = load_raw_mot(seq, mcmt_cfg) File "D:\THICV\Project\AIC21-MTMC\tracker\MOTBaseline\src\post_processing\main.py", line 17, in load_raw_mot mot_feat_dic = pickle.load(open(f'{mcmt_cfg.DATA_DIR}/{seq}/{seq}_mot_feat_raw.pkl', 'rb')) FileNotFoundError: [Errno 2] No such file or directory: 'D:/THICV/Porject/AIC21_MTMC/datesets/detect_provided/c042/c042_mot_feat_raw.pkl' mot_feat_dic = pickle.load(open(f'{mcmt_cfg.DATA_DIR}/{seq}/{seq}_mot_feat_raw.pkl', 'rb')) FileNotFoundError: [Errno 2] No such file or directory: 'D:/THICV/Porject/AIC21_MTMC/datesets/detect_provided/c046/c046_mot_feat_raw.pkl' sys args ars: ['main.py', 'c043', 'pp', 'aic_mcmt.yml'] using post processing loading tracked file D:/THICV/Porject/AIC21_MTMC/datesets/detect_provided/c043/c043_mot.txt Traceback (most recent call last): File "D:\THICV\Project\AIC21-MTMC\tracker\MOTBaseline\src\post_processing\main.py", line 110, in eval_seq(sys.argv[1], pp=sys.argv[2], mcmt_cfg=cfg) File "D:\THICV\Project\AIC21-MTMC\tracker\MOTBaseline\src\post_processing\main.py", line 41, in eval_seq results = load_raw_mot(seq, mcmt_cfg) File "D:\THICV\Project\AIC21-MTMC\tracker\MOTBaseline\src\post_processing\main.py", line 17, in load_raw_mot mot_feat_dic = pickle.load(open(f'{mcmt_cfg.DATA_DIR}/{seq}/{seq}_mot_feat_raw.pkl', 'rb')) FileNotFoundError: [Errno 2] No such file or directory: 'D:/THICV/Porject/AIC21_MTMC/datesets/detect_provided/c043/c043_mot_feat_raw.pkl' Traceback (most recent call last): File "D:\THICV\Project\AIC21-MTMC\reid\reid-matching\tools\trajectory_fusion.py", line 53, in cid_bias = parse_bias(cfg.CID_BIAS_DIR, scene_name) File "D:\THICV\Project\AIC21-MTMC\reid\reid-matching\tools\trajectory_fusion.py", line 30, in parse_bias with open(opj(timestamp_dir, sname + '.txt')) as f: FileNotFoundError: [Errno 2] No such file or directory: 'D:/THICV/Porject/AIC21_MTMC/datasets/AIC22_Track1_MTMC_Tracking/cam_timestamp/S06.txt' Traceback (most recent call last): File "D:\THICV\Project\AIC21-MTMC\reid\reid-matching\tools\sub_cluster.py", line 143, in for pkl_path in os.listdir(fea_dir): FileNotFoundError: [WinError 3] ϵͳ▒Ҳ▒▒▒ָ▒▒▒▒·▒▒▒▒: './exp/viz/test/S06/trajectory/' Traceback (most recent call last): File "D:\THICV\Project\AIC21-MTMC\reid\reid-matching\tools\gen_res.py", line 40, in map_tid = pickle.load(open('test_cluster.pkl', 'rb'))['cluster'] FileNotFoundError: [Errno 2] No such file or directory: 'test_cluster.pkl'

Dear authors,

Thank you very much on your great work!

I am a phd student from UIUC. Recently I am doing a research project related to traffic cam object tracking. I am wondering whether it's possible to provide instructions on how to run your model on training data S01 please? Specifically, we added or deleted some frames from the original videos from S01 to sync them up. Is there a way to run the model on our modified data of S01?

Thank you very much!

Hi,

Thank you for sharing this code, but I need to test this code and understand how does work for that I need a dataSet, can you give me small subset just for testing i.e understand your idea ?

thank you in advance