koehlp / wda_tracker Goto Github PK
View Code? Open in Web Editor NEWWeighted distance aggregation multi person multi camera tracker
License: MIT License
Weighted distance aggregation multi person multi camera tracker
License: MIT License
Hi @koehlp
Thanks a lot for the open repository. This is a great help!
I am using your trained models on a different sample dataset for inference. I got the SCT and now I want to do the clustering of the tracks across cameras. When I configure the configs/clustering_configs/mta_es_abd_non_clean.py
, do I need the "train_data" paths? Can I just perform a simple inference (as I don't have the gt values for my dataset)? If so, can you please help me as to what needs to be done?
Awesome work! I liked it :). How can i get an online MTMC tracking method? What are the needed changes? Do you have some useful code about it ?
Thanks in advance
Traceback (most recent call last):
File "multicam_trackwise_evaluation.py", line 383, in
result = Multicam_trackwise_evaluation(dataset_folder="/home/mca/Downloads/wda_tracker-master/MTA_ext_short/test"
File "multicam_trackwise_evaluation.py", line 195, in evaluate
track_eval_res_df = self.get_track_eval_res_df(summary)
File "multicam_trackwise_evaluation.py", line 275, in get_track_eval_res_df
idx_hids = id_global_assignment["idx_hids"]
KeyError: 'idx_hids'
Hi Philipp,
First of all, thanks for the excellent work!
I have encountered an "EOFError: Ran out of input" when running the command "run_multi_cam_clustering.py --config configs/clustering_configs/mta_es_abd_non_clean.py". This error was raised when calculating the pickled person_id_tracks. Below is the error message:
`Did not find pickled person_id_tracks. Calculating them now.
36%|███▋ | 423/1165 [00:29<02:17, 5.40it/s]Ran out of input
multiprocessing.pool.RemoteTraceback:
"""
Traceback (most recent call last):
File "/home/lxfhfut/anaconda3/envs/wda/lib/python3.7/multiprocessing/pool.py", line 121, in worker
result = (True, func(*args, **kwds))
File "/home/lxfhfut/Dropbox/PyCharm/wda_tracker/clustering/multi_cam_clustering.py", line 1413, in cluster_from_weights_task
, dataset_type=dataset_type
File "/home/lxfhfut/Dropbox/PyCharm/wda_tracker/clustering/multi_cam_clustering.py", line 1113, in cluster_from_weights
,dataset_type=dataset_type)
File "/home/lxfhfut/Dropbox/PyCharm/wda_tracker/clustering/multi_cam_clustering.py", line 818, in cluster_tracks_via_hierarchical
self.get_all_tracks_with_feature_mean(track_results_folder,dataset_type)
File "/home/lxfhfut/Dropbox/PyCharm/wda_tracker/clustering/multi_cam_clustering.py", line 239, in get_all_tracks_with_feature_mean
, dataset_type=dataset_type)
File "/home/lxfhfut/Dropbox/PyCharm/wda_tracker/clustering/multi_cam_clustering.py", line 280, in calculate_track_feature_mean
feature_dict = pickle.load(handle)
EOFError: Ran out of input
"""
The above exception was the direct cause of the following exception:
Traceback (most recent call last):
File "/home/lxfhfut/Dropbox/PyCharm/wda_tracker/run_multi_cam_clustering.py", line 108, in
run_clustering.run()
File "/home/lxfhfut/Dropbox/PyCharm/wda_tracker/run_multi_cam_clustering.py", line 84, in run
, n_split_parts=self.cfg.cluster_from_weights.split_count
File "/home/lxfhfut/Dropbox/PyCharm/wda_tracker/clustering/multi_cam_clustering.py", line 1377, in splitted_clustering_from_weights
eval_results = get_async_tracking_results(eval_results)
File "/home/lxfhfut/Dropbox/PyCharm/wda_tracker/clustering/multi_cam_clustering.py", line 1277, in get_async_tracking_results
result = result.get()
File "/home/lxfhfut/anaconda3/envs/wda/lib/python3.7/multiprocessing/pool.py", line 657, in get
raise self._value
EOFError: Ran out of input
`
I checked the '*/work_dirs/clustering/config_runs/mta_es_abd_non_clean/person_id_tracks' folder, nothing was generated.
Would you kindly help with this issue? Please let me know if you need further information to resolve the error. Thank you!
Hi Philipp,
Thank you very much for the excellent network.
I have some problems with visualizing multi-camera tracks. I run a file in utilities which named "draw_multi_camera_tracks.py" after the command “python run_multi_cam_clustering.py --config configs/clustering_configs/mta_es_abd_non_clean.py”
But there is an error that can't find the file which path is "work_dirs/evaluation/multi_cam_trackwise_evaluation/eval_results.csv"
Below is the error message:
Traceback (most recent call last):
File "draw_multi_camera_tracks.py", line 346, in
trv.run_visualization()
File "draw_multi_camera_tracks.py", line 284, in run_visualization
track_evaluation_results = self.read_track_evaluation_results()
File "draw_multi_camera_tracks.py", line 74, in read_track_evaluation_results
track_evaluation_results = pd.read_csv(self.track_evaluation_results_path)
File "/home/xb/anaconda3/envs/wda/lib/python3.7/site-packages/pandas/io/parsers.py", line 676, in parser_f
return _read(filepath_or_buffer, kwds)
File "/home/xb/anaconda3/envs/wda/lib/python3.7/site-packages/pandas/io/parsers.py", line 448, in _read
parser = TextFileReader(fp_or_buf, **kwds)
File "/home/xb/anaconda3/envs/wda/lib/python3.7/site-packages/pandas/io/parsers.py", line 880, in init
self._make_engine(self.engine)
File "/home/xb/anaconda3/envs/wda/lib/python3.7/site-packages/pandas/io/parsers.py", line 1114, in _make_engine
self._engine = CParserWrapper(self.f, **self.options)
File "/home/xb/anaconda3/envs/wda/lib/python3.7/site-packages/pandas/io/parsers.py", line 1891, in init
self._reader = parsers.TextReader(src, **kwds)
File "pandas/_libs/parsers.pyx", line 374, in pandas._libs.parsers.TextReader.cinit
File "pandas/_libs/parsers.pyx", line 674, in pandas._libs.parsers.TextReader._setup_parser_source
FileNotFoundError: [Errno 2] File /home/xb/wda_tracker/wda_tracker-master/work_dirs/evaluation/multi_cam_trackwise_evaluation/eval_results.csv does not exist: '/home/xb/wda_tracker/wda_tracker-master/work_dirs/evaluation/multi_cam_trackwise_evaluation/eval_results.csv'
Should I run some file to create "eval_results.csv" before running "draw_multi_camera_tracks.py"?
Or did I run the wrong file to visualize multi-camera tracks?
Look forward to your reply, thank you!
what I have to say is, this code is hard to run, when I run the tracker "run_tracker.py", especially the detector module
Hi,
I met some questions when I install python requirements.
I can't find the corresponding version of mkl-fft==1.0.15 and mkl-random==1.1.0
so i install them with version==1.2.0, but they are incompatible with the numpy's version==1.18.1
so i upgrade the version of numpy to 1.19.5, then run run_tracker.py met the following problem
could you please tell me how to modify this problem?
Nice work! I want to train model in another game dataset, could you provide the training code?
Hi,
I do some modifications on the original code and when I want to do the evaluation, I meet key Error in metrics.py
I think there may be some problem and could u provide your cam tracking results? I know that you have provide the multi-cam results in other issues. I replace my files with them but met some errors.
Thanks.
Hi,
Thanks for the open repository. it is a great project!
I want to use the WDA model to test on my own dataset, so how can I modify the project?
Thanks
Hi,
Thanks for the open repository. it is a great project!
I had completed Run the single camera tracking and Run the multi camera clustering, got some revelant files.
so what need to do if i want to visualize the tracks?
Thanks
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.